Flinders University - IMPROVING E-GOVERNMENT ...flex.flinders.edu.au/.../1/Thesis_Hanafiah_2016.pdfM.Sc. (Comp. Info. Systems) Waseda University, Japan B.Sc. (Comp. Engineering) Gunadarma

IMPROVING E-GOVERNMENT PERFORMANCE

THROUGH ENTERPRISE ARCHITECTURE IN

DEVELOPING COUNTRIES:

THE CASE OF THE INDONESIAN TREASURY

By

Mochamad Ali Hanafiah

M.Sc. (Comp. Info. Systems) Waseda University, Japan

B.Sc. (Comp. Engineering) Gunadarma University, Indonesia

School of Computer Science, Engineering and Mathematics

Faculty of Science and Engineering

2015

A thesis presented to the

Flinders University of South Australia

In fulfilment of the requirements for the degree of

Doctor of Philosophy

Adelaide, South Australia, 2015

© Mochamad Ali Hanafiah

i

CERTIFICATION

I certify that this thesis does not incorporate without acknowledgment any material

previously submitted for a degree or diploma in any university; and that to the best of

my knowledge and belief it does not contain any material previously published or

written by another person except where due reference is made in the text.

Signed Dated

Mochamad Ali Hanafiah 31 July 2015

ii

Table of Contents

Certification ........................................................................................................ i

List of Tables .................................................................................................... xi

List of Figures ................................................................................................. xiv

Abstract ...................................................................................................... xvii

Publications since PhD enrolment ................................................................ xix

List of Abbreviations ...................................................................................... xx

Acknowledgement .......................................................................................... xxi

Introduction .................................................................................... 1

1.1 Overview ................................................................................................... 1

1.2 Research Background ............................................................................... 1

1.3 Statement of the problem .......................................................................... 3

1.4 Research Aims and Questions................................................................... 6

1.5 Methodology ............................................................................................. 7

1.6 Research significance ................................................................................ 8

1.7 Thesis outline .......................................................................................... 11

1.8 Conclusion and Summary ....................................................................... 13

Literature Review ........................................................................ 14

2.1 Overview ................................................................................................. 14

2.2 E-Government ......................................................................................... 14

2.2.1 The evolution of e-Government .............................................................. 14

2.2.2 Definition of e-Government .................................................................... 17

iii

2.2.3 Implication of e-Government .................................................................. 18

2.2.4 e-Government in Developing Countries ................................................. 20

2.2.5 Technology Adoption Perspective of e-Government .............................. 20

2.2.6 Developing and Deploying e-Government ............................................. 24

2.2.7 Achieving Strategic Values of e-Government ........................................ 25

2.2.8 Measuring e-Government Performance .................................................. 26

2.3 Enterprise Architecture ........................................................................... 30

2.3.1 The Evolution of Enterprise Architecture ............................................... 31

2.3.2 Definition of Enterprise Architecture...................................................... 34

2.3.3 Aligning Business and Technology by using Enterprise Architecture ... 36

2.3.4 The Use of Enterprise Architecture and Other Instruments .................... 36

2.3.5 Measuring the Progress of Enterprise Architecture ................................ 38

2.4 Enterprise Architecture Contribution to e-Government in Developed

Countries ................................................................................................. 39

2.4.1 E-Government Development Index ........................................................ 40

2.4.2 EA in Developed Countries .................................................................... 43

2.4.3 Significance of EA to e-Government ...................................................... 45

2.5 EA in Developing countries .................................................................... 47

2.6 Asian Countries’ Culture Characteristics ................................................ 48


Research Model ............................................................................ 50

3.1 Overview ................................................................................................. 50

3.2 The relationship between EA and e-Government ................................... 50

iv

3.3 Enterprise Architecture Benefits ............................................................. 51

3.4 Measuring e-Government Performance using SERVQUAL .................. 54

3.5 Developing an Alternative EA Maturity Model ...................................... 56

3.5.1 The meta-synthesis approach .................................................................. 56

3.5.2 Meta-synthesis research process ............................................................. 57

3.6 Refining the EABM ................................................................................ 65

3.6.1 Mapping EABM’s benefits and e-Government performance’s variables 66

3.6.2 Proposed Model ...................................................................................... 68

3.7 Hypotheses .............................................................................................. 71

3.8 Definition of Constructs .......................................................................... 73


Research Methodology ................................................................ 76

4.1 Overview ................................................................................................. 76

4.2 Case Study Research ............................................................................... 76

4.3 Research framework ............................................................................... 78

4.4 Identification of the Data Sources ........................................................... 80

4.5 Selection of Methods: Mixing Qualitative and Quantitative Methods ... 81

4.6 Selection of Respondents ........................................................................ 83

4.7 Research Techniques............................................................................... 84

4.7.1 Interview ................................................................................................. 85

4.7.2 Questionnaire .......................................................................................... 86

4.7.3 Documentary Evidence ........................................................................... 88

4.7.4 Observations ........................................................................................... 89

v

4.8 Data collection, Coding and Analysis ..................................................... 90

4.9 Ethical Issues .......................................................................................... 92


The Indonesian Treasury e-Government ................................... 94

5.1 Overview ................................................................................................. 94

5.2 The Core of The Indonesian Public Financial Systems .......................... 94

5.3 The Antecedent of e-Government Systems in the Indonesian Treasury . 97

5.3.1 The Systems Evolution ........................................................................... 97

5.3.2 Data Communication .............................................................................. 99

5.4 E-Government Systems in the Indonesian Treasury ............................. 102

5.4.1 Software Development Life Cycle ........................................................ 102

5.4.2 The Systems .......................................................................................... 103

5.5 ICT Organisation Evolution .................................................................. 105

5.6 ICT Resources ....................................................................................... 108

5.6.1 Hardware and Software ......................................................................... 108

5.6.2 Human Resources ................................................................................. 111

5.7 Conclusion and Summary ..................................................................... 112

Qualitative Analysis ................................................................... 113

6.1 Overview ............................................................................................... 113

6.2 Phases of Analysis ................................................................................ 113

6.3 Respondent Profiles .............................................................................. 114

6.4 Content Analysis ................................................................................... 118

6.4.1 All respondents ..................................................................................... 119

vi

6.4.2 Echelon 1 (2 Respondents) ................................................................... 119



6.4.5 Comparison ........................................................................................... 122

6.5 Thematic Analysis ................................................................................ 124

6.5.1 Enterprise Architecture ......................................................................... 125

6.5.2 Organisational Alignment ..................................................................... 127

6.5.3 Information Availability ....................................................................... 129

6.5.4 Resource Portfolio Optimisation ........................................................... 132

6.5.5 Resource Complementarity ................................................................... 133

6.5.6 E-Government Systems Performance ................................................... 134

6.6 Documentary Evidence ......................................................................... 136

6.6.1 Government Regulations....................................................................... 136

6.6.2 Administration Mailing Systems .......................................................... 136

6.7 Observation ........................................................................................... 137


Quantitative Findings ................................................................ 139

7.1 Overview ............................................................................................... 139

7.2 Developing the Measures ...................................................................... 139

7.3 Validity Test of the Measures ............................................................... 142

7.3.1 Panel of Experts .................................................................................... 143

7.3.2 Pre-Test ................................................................................................. 144

7.4 Subject and Data Collection .................................................................. 144

vii

7.5 Descriptive Statistics of Respondents Data .......................................... 145

7.6 Data Screening ...................................................................................... 149

7.6.1 Handling Missing Data ......................................................................... 149

7.6.2 Preliminary Data Analyses.................................................................... 151

7.7 Data Reliability, Validity and Bias Tests .............................................. 152

7.7.1 Construct Validity Test ......................................................................... 152

7.7.2 Reliability Test ...................................................................................... 162

7.7.3 Common Method Variance Bias ........................................................... 163

7.8 The goodness-of-fit ............................................................................... 164

7.9 Generating and Validating the Model Dimensions ............................... 167

7.9.1 The Unchanged Dimensions ................................................................. 168

7.9.2 Resource Portfolio Optimisation ........................................................... 168

7.9.3 e-Government Performance .................................................................. 170

7.10 Generating and Validating the Overall Model ...................................... 172

7.11 Mediator Analysis of the Model ........................................................... 177

7.11.1 An Overview of Simple Mediation Analysis using SEM ..................... 177

7.11.2 Results for Mediation Analysis of the Model ....................................... 178


Discussion .................................................................................... 183

8.1 Overview ............................................................................................... 183

8.2 EA contributions to e-Government ....................................................... 183

8.3 The Effect of EA on e-Government ...................................................... 183

8.4 The Presence of EA in the Indonesian Treasury ................................... 184

viii

8.5 The Indonesian Treasury e-Government Performance ......................... 185

8.6 Answers to Hypotheses ......................................................................... 186

8.6.1 H1: Organisational Alignment Mediates the Positive Effect of EA on e-

Government Performance ..................................................................... 187

8.6.2 H2: Information Availability Mediates the Positive Effect of EA on e-


8.6.3 H3: Resource Portfolio Optimisation Mediates the Positive Effect of EA

on e-Government Performance ............................................................. 192

8.6.4 H4: Resource Complementarity Mediates the Positive Effect of EA on e-



Conclusions and Further Research .......................................... 196

9.1 Overview ............................................................................................... 196

9.2 Research Aims and Objectives ............................................................. 196

9.3 Summary of the results ......................................................................... 197

9.4 Answers to Research Questions ............................................................ 200

9.5 Contributions of the Study .................................................................... 202

9.6 Policy Implications ............................................................................... 205

9.7 Limitations and Further Research ......................................................... 206


References ...................................................................................................... 209

Ethics Approval, Letter Of Introduction, Invitation To

Research, And Consent Form ...................................................................... 228

Questionnaire ....................................................................... 240

ix

List of Interview Questions ................................................. 249

List of Regulations ............................................................... 251

SQL Query Syntax ............................................................... 252

Extracts from Mailing Administration Systems ................ 253

The Indonesian Treasury Organisation ............................ 255

The Indonesian Treasury e-Government Systems Basic

Configuration ............................................................................................... 258

Descriptive Statistics of The Observed Variables .............. 263

Enterprise Architecture .......................................................................................... 263

Organisational Alignment...................................................................................... 263

Information Availability ........................................................................................ 263

Resource Portfolio Optimisation ........................................................................... 264

Resource Complementarity ................................................................................... 264

e-Government Performance using Initial SERVQUAL Congeneric Models ........ 264

Tangible Factor ...................................................................................................... 264

Reliability Factor ................................................................................................... 265

Responsiveness Factor ........................................................................................... 265

Assurance Factor ................................................................................................... 265

Empathy Factor ..................................................................................................... 266

Covariance Models For Validity Tests ............................... 267

Structural Equation Models ............................................... 269

E-Government Dimension ..................................................................................... 269

The Overall Model ................................................................................................ 270

x

The Simple Mediator Models ................................................................................ 272

Statistical Equations ............................................................ 276

CHI-SQUARE (χ2) ................................................................................................ 276

DEGREES OF FREEDOM (df) ............................................................................ 276

SRMR .............................................................................................................. 276

RMSEA .............................................................................................................. 276

CFI .............................................................................................................. 276

IFI .............................................................................................................. 276

TLI .............................................................................................................. 276

xi

LIST OF TABLES

Table 2.1. Shifting from bureaucracy to e-Government (Tat-Kei Ho, 2002) ............ 19

Table 2.2. E-Government Stages ................................................................................ 27

Table 2.3. EA maturity models .................................................................................. 39

Table 2.4. Culture Characteristics (Trompenaars & Woolliams, 2011) ..................... 49

Table 3.1. EA Maturity model comparison ................................................................ 58

Table 3.2. Comparison Among Maturity Models ...................................................... 60

Table 3.3. Illustrating relationship between one model to another ............................ 62

Table 3.4. The six As EA maturity model .................................................................. 64

Table 3.5. Definition of Constructs ............................................................................ 74

Table 4.1. Purpose of Mixed Methods (Venkatesh et al., 2013, p. 26) ...................... 82

Table 4.2. Respondent Selection Technique .............................................................. 83

Table 4.3. Research Techniques and Sources of Data ............................................... 85

Table 4.4. Survey Response Rate ............................................................................... 88

Table 4.5. Dimension Reduction of Observed Variables ........................................... 91

Table 6.1. Interviewee Role and Interview Date ...................................................... 115

Table 6.2. Response to Interview Request and Respondent ID based on ranks ....... 116

Table 6.3. Age distributions ..................................................................................... 116

Table 6.4. Respondent’s Location ............................................................................ 117

Table 6.5. Respondent’s Role ................................................................................... 117

Table 6.6. Educational background .......................................................................... 117

Table 6.7. Respondent’s educational background .................................................... 117

Table 6.8. Comparison of Most Frequently Used Words by All Respondents ........ 122

Table 7.1. List of Questions/Statements Developed ................................................ 141

Table 7.2. List of Questions/Statements adopting SERVQUAL ............................. 142

xii

Table 7.3. Group of validity responses received from experts ................................. 144

Table 7.4. Respondent Gender ................................................................................. 146

Table 7.5. Respondent Group of Age in years old ................................................... 146

Table 7.6. Working Period ....................................................................................... 147

Table 7.7. Working place ......................................................................................... 147

Table 7.8. Office Location Area ............................................................................... 148

Table 7.9. Role in Office .......................................................................................... 148

Table 7.10. Educational Background ....................................................................... 149

Table 7.11. Total Missing Responses for Observed Variables ................................ 150

Table 7.12. Convergent Validity Analysis for the e-Government Dimension ......... 154

Table 7.13. Factor Score Weight .............................................................................. 155

Table 7.14. EFA Results Comparison ...................................................................... 156

Table 7.15. Convergent Validity of the Measures .................................................... 159

Table 7.16. Correlation Between Factors ................................................................. 161

Table 7.17. Correlation Between Items and Factors ................................................ 161

Table 7.18. Cronbach Alpha and Mean Inter-Item Correlations .............................. 163

Table 7.19. Harman's Single Factor Test ................................................................. 164

Table 7.20. The goodness-of-fit cut-off values (Barrett, 2007; Byrne, 2010; Hu &

Bentler, 1999; Iacobucci, 2010; Marsh et al., 2004; Susanto, 2012) ....................... 166

Table 7.21. The GOF of Research Model Dimensions ............................................ 167

Table 7.22. RP Modification Indices - Covariances ................................................ 169

Table 7.23. e-Gov Dimension Modification Indices – Covariances ........................ 171

Table 7.24. The GOF for Overall Model .................................................................. 173

Table 7.25. Overall Modification Indices – Covariances ......................................... 174

Table 7.26. The GOF for Mediation Analysis .......................................................... 178

xiii

Table 7.27. The Effect of Each Mediator on e-Government Performance .............. 180

Table 7.28. The Significance of Mediator Effect ..................................................... 181

Table 9.1. Final SERVQUAL Congeneric Models .................................................. 204

Table A. 1. The Number of Request for Reports ..................................................... 253

Table A. 2. The Number of Request for Data .......................................................... 253

Table A. 3. The Number of Request for Data or Report Reconcilation ................... 254

Table A. 4. Main Systems Used for Treasury Function ........................................... 258

xiv

LIST OF FIGURES

Figure 1.1. The Structure of The Thesis ..................................................................... 12

Figure 2.1. Technology Acceptance Model (TAM) (Davis et al., 1989, p. 985) ....... 21

Figure 2.2. The Unified Theory of Acceptance and Use of Technology (UTAUT)

(Venkatesh et al., 2003, p. 447) .................................................................................. 22

Figure 2.3. The Technology-Organisation-Environment (TOE) (Baker, 2012, p. 236;

Depietro et al., 1990, p. 153) ...................................................................................... 23

Figure 2.4. The complete Zachman Framework for Enterprise Architecture (Zachman,

2003 cited in Finkelstein, 2006, p. 4) ......................................................................... 33

Figure 2.5. Management areas related to Enterprise Architecture (Lankhorst, 2009, p.

14) ............................................................................................................................... 37

Figure 2.6. Overall e-Government Index among observed countries ........................ 45

Figure 2.7. Web Measure/Online Service Index ........................................................ 46

Figure 3.1. The EABM (Tamm et al., 2011) .............................................................. 53

Figure 3.2. Mapping EABM’s Benefit Enablers and SERVQUAL variables ........... 68

Figure 3.3. Relationship of EA and e-Government Performance .............................. 70

Figure 3.4. Hyphothesis 1 of the study ....................................................................... 72




Figure 4.1. The Research Process Of This Study ....................................................... 77

Figure 4.2. Literature Coverage in This Study ........................................................... 79

Figure 5.1. The role of the Indonesian Treasury e-Government Systems ................. 96

Figure 5.2. Systems Evolution in the Indonesian Treasury ........................................ 98

xv

Figure 5.3. Telephone, Internet and Fixed Broadband subscribers in Indonesia per 100

people (The World Bank, 2014) ............................................................................... 100

Figure 5.4. SDLC Framework .................................................................................. 103

Figure 5.5. Separation of Treasury Function ............................................................ 104

Figure 5.6. ICT Organisation Evolution ................................................................... 106

Figure 5.7. Total Computer Assets as of December 2013 ........................................ 109

Figure 5.8. Operating Systems Used as of 2008 ...................................................... 109

Figure 5.9. Education Background of all staff in the Indonesian Treasury .............. 111

Figure 6.1. Five-phase cycle (Yin, 2011, p. 178) ..................................................... 114

Figure 6.2. Most Frequently Used Words by All Respondents ................................ 119

Figure 6.3. Most Frequently Used Words by Echelon 1 .......................................... 120



Figure 6.6. Radar View of Frequent Words Used .................................................... 124

Figure 6.7. The growth of requests from the Indonesian Treasury Headquarters .... 137

Figure 7.1. Quantitative Analysis Flowchart ............................................................ 140

Figure 7.2. The Relationship of Latent Variables in the Model ............................... 153

Figure 7.3. Modified e-Government Performance Dimension ................................. 158

Figure 7.4. EA and OA Standarised Estimates ........................................................ 168

Figure 7.5. IA and RC Standarised Estimates .......................................................... 168

Figure 7.6. Initial RP Standarised Estimates ........................................................... 168

Figure 7.7. Modified RP Standarised Estimates ....................................................... 170

Figure 7.8. Final e-Gov Dimension Standarised Estimates...................................... 172

Figure 7.9. Standarised Estimates for the Final Structural Equation Model ............ 176

xvi

Figure 7.10. Conceptual Diagram of Simple Mediation Model (Hayes, 2013, p. 87;

Iacobucci, 2008, p. 2) ............................................................................................... 177

Figure 7.11. Direct Effect of EA on e-Government without mediators ................... 180

Figure 8.1. The Effect of OA as Mediator ............................................................... 187

Figure 8.2. The Effect of IA as Mediator ................................................................. 190

Figure 8.3. The Effect of RP as Mediator ................................................................ 192

Figure 8.4. The Effect of RC as Mediator ................................................................ 194

Figure A. 1. Job Rank Distribution .......................................................................... 255

Figure A. 2. The Indonesian Treasury Staff Age ..................................................... 256

Figure A. 3. Organisational Structure of Treasury Regional Office ........................ 256

Figure A. 4. Organisational Structure of Treasury Service Office Type A1 ............ 257

Figure A. 5. Organisational Structure of Treasury Service Office Type A2 ............ 257

Figure A. 6. Initial e-Government Performance Covariance Models ...................... 267

Figure A. 7. Covariance Model for All Dimensions ................................................ 268

Figure A. 8. Initial e-Government Standarised Estimates ........................................ 269

Figure A. 9. Standarised Estimates for the Initial Model ......................................... 270

Figure A. 10. Standarised Estimates for the Final Overall Model ........................... 271

Figure A. 11. Standarised Estimates for the OA Dimension as Mediator ................ 272

Figure A. 12. Standarised Estimates for the IA Dimension as Mediator ................. 273

Figure A. 13.Standarised Estimates for the RP Dimension as Mediator ................. 274

Figure A. 14. Standarised Estimates for the RC Dimension as Mediator ................ 275

xvii

ABSTRACT

This study investigated ways of improving e-Government performance in

developing countries. Developing countries were selected as they have significant

barriers to the development of sustainable e-Government systems. A review of the

literature found that improving e-Government performance requires more than just

technology solutions. Instead, a holistic view of e-Government is needed. This can be

achieved through Enterprise Architecture (EA).

EA is defined as “a blueprint that documents the information systems within the

enterprise, their relationships, and how they interact to fulfil the enterprise’s mission”

(Langenberg & Wegmann, 2004, p. 2). While governments of developed countries

have used EA, it has not yet been fully utilised by developing countries. Despite recent,

growing interest in EA from developing countries, empirical data to show the effect of

EA adoption on e-Government performance is limited.

This study builds on the EA Benefit Model (EABM) from Tamm, Seddon,

Shanks, and Reynolds (2011) and extends it by using SERVQUAL to measure e-

Government performance. This modified EABM is tested in a case study of a

developing country, Indonesia, and validated in the context of other developing

countries. As a result of this analysis, this study offers valuable examples of policy

implications for assisting developing countries, in general, and the Indonesian

government in particular. This is vital for Indonesia in order to address concerns

identified in the United Nations e-Government readiness reports since 2003.

This study employed mixed methods with two different groups of respondents.

A quantitative approach, involving a survey of 561 respondents, was used to collect

data from respondents working at operational levels. A qualitative phase involving

xviii

fifteen semi-structured interviews was conducted with high-level officials who have

the ability to influence the strategic decisions relating to e-Government systems

development. In addition, documentary evidence and observations were used to

capture a more comprehensive picture of the Indonesian Treasury e-Government

systems.

Using Structural Equation Modelling (SEM) in Amos version 22, the modified

EABM and the collected data supports all hypotheses made in this study. The findings

revealed that EA has the potential to improve e-Government system performance in

Indonesia. Also, the study developed the SERVQUAL measures further to address

instability problems (Jiang et al., 2002; Landrum et al., 2009; Myerscough, 2002)

resulting in a validated and stable measure.

Therefore, this study provides an important theoretical contribution to the e-

Government literature in advancing understanding of the critical role of Enterprise

Architecture in improving the quality of e-Government systems and in measuring e-

Government performance in developing countries.

xix

PUBLICATIONS SINCE PHD ENROLMENT

1. Hanafiah, M. A., & Goodwin, R. (2011). Government Enterprise Architecture:

lessons learned from the United States’ FEAF and Australia’s AGA.

International Information Systems Conference (ISICO), 1, 97-102.

2. Hanafiah, M. A., & Goodwin, R. (2012). An Alternative Enterprise

Architecture Maturity Model. IJACP-Proceedings of the International

Conference on Business Management & Information Systems, 509-517.

3. Hanafiah, M. A., & Goodwin, R. (2013, 2-4 December 2013). Does Enterprise

Architecture give value to e-Government in Developed Countries? Paper

presented at the Information Systems International Conference (ISICO), Bali,

Indonesia.

4. Hanafiah, M. A., & Goodwin, R. (2014). Improving e-Government

Performance through Enterprise Architecture. In G. Rampersad & F. Patel

(Eds.), Technology Innovation Leadership in Development: A Middle East

(West Asia) Perspective (pp. 35-58). New York: Nova Publishers.

xx

LIST OF ABBREVIATIONS

AGA The Australian Government Architecture

CFA Confirmatory Factor Analysis

CIO Chief Information Officer

CMMI Capability Maturity Model Integration

COBIT Control Objectives for Information and Related Technology

COTS Commercial off-the shelf

DG Directorate General

DTS Directorate of Treasury Systems

DTT Directorate of Treasury Transformation

EA Enterprise Architecture

EABM The EA Benefit Model

EAP Enterprise Architecture Planning

EFA Exploratory Factor Analysis

EFQM The European Foundation for Quality Management

FEAF Federal Enterprise Architecture Framework

GOF The Goodness-of-Fit

ICT Information and Communication Technology

IFMIS Integrated Financial Management Information Systems

IS Information Systems

ISO The International Organization for Standardization

ITIL The Information Technology Infrastructure Library

ITU The International Telecommunication Union

LAN Local Area Network

MoF The Indonesian Ministry of Finance

NPM New Public Management

OECD Organisation for Economic Co-operation and Development

OS Operating System

PEM Public Expenditure Management

SDLC Software Development Life Cycle

SEM Structural Equation Modelling

SGEA Singapore Government Enterprise Architecture SPAN Sistem Perbendaharaan dan Anggaran Negara or State Budget and Treasury

Systems

TAM Technology Acceptance Model

TOE The Technology-Organisation-Environment

TOGAF The Open Group Architecture Framework

TRA Theory of Reason Action

TRO The Indonesian Treasury Regional Office (KANWIL)

TSO The Indonesian Treasury Service Office (KPPN)

UN United Nations

UTAUT The Unified Theory of Acceptance and Use of Technology

WAN Wide Area Network

xxi

ACKNOWLEDGEMENT

First and foremost, I would like to thank Allah, God Almighty, for his blessing,

and for giving me the physical and mental strength throughout the enlightening journey

of this study.

I would like to sincerely thank my principal supervisor Dr. Robert Goodwin for

giving me freedom to discover my interests and for his exceptional support,

encouragement, guidance and patience throughout my candidature. I would also like

to thank Dr. Giselle Rampersad, my associate supervisor, for her outstanding

contribution, invaluable comments, motivation and for considering every detail of the

study. Both of them instilled confidence to persistently carry out the research.

My special gratitude belongs to the Australian government for its generous aid

through a prestigious scholarship, the Australian Leadership Award (ALA). It

provided me with the incredible experience of conducting my PhD study at Flinders

University; the place where I met amazing and inspiring people who gave me

extraordinary opportunity to undertake teaching and math advisor roles: Associate

Professor Salah Kutieleh, the Director of Transition office, and Dr. Kathleen Brady,

Head of the Student Learning Centre.

I would also like to thank the Indonesian Ministry of Finance especially Mr.

K.A. Badaruddin, Mr. Herry Purnomo, Mr. Agus Suprijanto, Mr. Bobby A.A. Nazief,

Mr. Paruli Lubis, Ms. Sri Hartati, Mr. Sudarto, Mr. Eko Sulistijo, Mr. Achmad Rinaldi

Hidayat and colleagues who have supported me to complete this study. I am indebted

to all participants in the Indonesian Treasury who shared their priceless time to

contribute in this research.

xxii

I am so grateful to have supportive communities such as MIIAS, PBB, Marion

Masjid, PPIA, the Treasury Troopers, FUMA, KIA, AIA and IICSA who made my

life so fruitful in Adelaide. I got to know wonderful people who kept on supporting

me: the Cody family, the MacKenzie family, Dr. Darfiana Nur, Dr. Elvia Shauki, Dr.

Anthony Cramp, Indonesian PhD colleagues: Iwan Agung Prasetyo, Widi Nugroho,

Sukendar, Komari, Fadliadi, Budi Sunarko, all who endeavoured on similar journies,

and others that I could not mention individually in my very long list.

Finally, my heartfelt gratitude goes to my lovely, encouraging, and caring wife,

Rizka Yuniarsih, for her countless assistance, and to my kids, Kayla and Izzan for

energising me with enjoyable distractions from my research. For my mother, Rasimah

Rapiun, my brothers, Faisal, Imran, Hendry, and Ruhiyat, my sisters, Ririn, Dieni and

Fatia, your love and prayers made this happen. I dedicate this thesis to the souls of my

dad, my father in law and my mother in law. This is a family achievement.

1

INTRODUCTION

1.1 Overview

This study explores e-Government in developing countries and enterprise

architecture (EA). This chapter begins, in Section 1.2, by describing the background

of this study. It is followed by Section 1.3 discussing the problem motivating this

study. Section 1.4 provides the research aim and objective and also introduces the

research questions. Section 1.5 highlights the research methodology used in the study.

Section 1.6 proposes the significance of this study. The outline of this thesis is

available in Section 1.7. Section 1.8 summarises this chapter.

1.2 Research Background

Governments require reliable information to improve public services. The

information should be provided through a secure channel in a timely manner to remain

useful. Thus, the presence of Information and Communication Technology (ICT) to

provide reliable information systems is indispensable. “The use of ICT in government”

is defined as e-Government (OECD, 2003, p. 23). The OECD further states that e-

Government “encompasses all aspects of government activity”.

Unlike some governments of developed countries who are able to make the most

of e-Government, most governments of developing countries are still struggling to

implement proper e-Government systems. Heeks (2003, 2008) argues that only 15%

of e-Government projects in developing or transitional countries were considered

successful. Ebrahim and Irani (2005) suggest that to make e-Government successful,

government should not rely solely on the ICT.

2

With the rapid development of ICT, a successful information system such as e-

Government system entails strong and continuous support from all stakeholders (Irani,

2002). A collaborative environment is also needed in the government. In order to

provide such an environment, interoperability of data exchange is necessary.

Therefore, governments require an instrument that can be used by various stakeholders

to communicate in the same logical construct and to cope with the changes in the ICT.

Zachman (1997) argues that EA could be used for such circumstances.

EA is defined as a“blueprint that documents all the information systems within

the enterprise, their relationships, and how they interact to fulfil the enterprise’s

mission” (Langenberg & Wegmann, 2004, p. 2). Governments are considered to be

the most complex enterprises in the world (Saha, 2010a). Therefore, some

governments of developed countries have started to adopt EA in their e-Government

initiatives. The first effort came from the United States government (CIO Council,

1999).

The number of countries adopting EA as part of their e-Government strategies

is growing. Yet, this study could not find any developing countries trailing such

initiatives. Therefore, this study aims to discover whether the presence of EA would

enhance the e-Government performance in developing countries such as Indonesia.

Indonesia is one of the lowest scoring countries in the United Nations’ e-Government

readiness survey (The United Nations, 2012). In particular, Indonesian e-Government

systems are considered to be lacking in their support for financial transactions (The

United Nations, 2012, 2014).

The Indonesian Treasury e-Government system is the host for government

financial related transactions such as expenditure, revenue, and accounting. Layne and

3

Lee (2001, p. 125) argue that “the critical benefits of implementing e-Government are

actually derived from the integration of underlying processes”. Hence, an in-depth

analysis of the Indonesian Treasury e-Government systems is deemed important as it

can enrich the e-Government field of research (Walsham & Sahay, 2006).

1.3 Statement of the problem

Developing countries are considered to have difficulties in promoting

transparency in their fiscal information (Kopits & Craig, 1998). E-Government is

considered to be a major contributor in improving transparency (Bertot et al., 2010;

Ciborra, 2005; OECD, 2003). In Indonesia, fiscal information is managed in the

Ministry of Finance with the Indonesian Treasury at the heart of its process (Achmad,

2012). This leads the Indonesian Treasury e-Government system to be critical for the

country. Together with other governments such as Mongolia, Turkey, Guatemala, and

Pakistan, Indonesia was part of World Bank projects to enhance their treasury e-

Government systems (Dener et al., 2010).

Despite assistance from international donors such as the World Bank, the

Indonesian Treasury received ongoing recommendations regarding its e-Government

systems from the Indonesian supreme audit board: Badan Pemeriksa Keuangan

(BPK)1. In their reports from 2005 to 2013, it was suggested that the Indonesian

Treasury should have a consistent architectural framework from which its e-

Government systems are developed (Badan Pemeriksa Keuangan, 2009, 2014). BPK

argued that the system was not developed upon relevant information system

1 The Indonesian supreme audit board is called Badan Pemeriksa Keuangan (BPK) in the

Indonesian language.

4

architectural patterns. This resulted in lack of assurance in the quality and reliability

of the Indonesian government financial report.

An architectural pattern like EA could be used for the Indonesian Treasury e-

Government systems. As a core government agency in public financial management,

the Indonesian Treasury needs to ensure that its systems are interoperable with other

agencies. This is required so that business and technology personnel can communicate

meaningfully in the same “language”. Referring to other mature industries such as

airlines and construction, Zachman and Sowa (1992) argue that EA can capture all the

processes for ICT development. Hence, using EA allows all parties, including business

and technology staff, involved in developing information systems to refer to the same

resource.

The Indonesian Treasury e-Government system is used to serve the Indonesian

government spending units through 177 Treasury Service Offices (TROs) and 30

Treasury Regional Offices (TROs) nationwide. The locations of the spending units are

widely dispersed. They are located in main cities, rural areas and some in remote areas.

Some of them are also located in foreign countries, such as embassies and consulates.

To achieve its complete role, the Indonesian Treasury needs to gather information from

all these, and other, sources in a timely manner.

To date, the Indonesian Treasury e-Government system is represented by more

than twenty system applications that are used for Treasury services, both in the

Indonesian Treasury and the government spending units. The number of system

applications and other software is still growing. The growth follows its user’s needs.

These systems are internally developed by more than fifty system developers with

various skills within the Indonesian Treasury. The system applications implement

5

services such as budget allotment, disbursement, cash management, accounting and

reporting. Each of the applications has more than five different sub-system

applications.

The large number of system applications leads to an increase in the complexity

of development (Basili & Perricone, 1984) and maintenance (Banker et al., 1993).

Furthermore, the Indonesian Treasury e-Government systems are developed using

multiple platforms. Most of them are developed by using Microsoft Visual FoxPro.

Each of the system applications can be run either on a stand-alone computer or used

in a client-server configuration over a Local Area Network (LAN).

The need for system applications to be installed on each client computer means

that data integrity is vitally important. The Indonesian Treasury’s business processes

are interdependent and data integration is managed through a periodic synchronisation

process. As such, each system application is equipped with a data transfer feature. This

feature is run regularly both manually and semi-automatically.

As such, the Indonesian Treasury has faced difficulties in their e-Government

initiative. The state of the Indonesian Treasury e-Government systems is similar to

what Peristeras and Tarabanis (2000), cited in Hjort-Madsen (2006, p. 2), described:

The existence of isolated, overlapping, highly fragmented, and unrelated

computerized applications within the same organization has resulted in major

interoperability problems and ‘isolated islands of technology’.

Therefore, one of the Treasury missions, i.e., to have an integrated, transparent,

accurate and real time information system (The Indonesian Treasury, 2007), is yet to

be achieved.

Investigations are needed on the Indonesian Treasury in order to find out whether

design has an impact on its e-Government system performance (Dada, 2006). The

6

design should be viewed as a combination of technological and business perspectives.

Hence, this study should be able to discover the main problems and how a sustainable

solution is formulated. Additionally, such a study would be greatly useful for donor

organisations such as the World Bank in their relentless efforts to assist developing

countries (Dener et al., 2010).

1.4 Research Aims and Questions

The aim of this study is to investigate ways to improve e-Government

performance in developing countries. This required a thorough analysis of government

documents such as laws and regulations. Accessing government documents in

developing countries, according to Poulin (2004), is considered difficult. However, the

researcher’s background as a member of the staff in the Indonesian Treasury helped in

accessing information that is not publicly available. Additionally, to the best of the

researcher’s knowledge, no similar study has been undertaken.

In order to achieve the aims of this study, relevant instrumentation is needed to

measure how the e-Government systems perform and how they could be improved.

The e-Government system performance cannot simply be measured by adoption of the

latest available technology. Also required are the views of the Indonesian Treasury

employees’ satisfaction regarding the current e-Government systems. The study also

needs to find out how the Indonesian Treasury developed its e-Government systems.

The presence of alignment between technology and business in the Indonesian

Treasury is also examined.

EA aims for alignment of technology and business. EA has gained attention not

only in private enterprise but also government enterprises. Governments have been

starting to adopt EA for their e-Government programs. Although EA is recognized as

7

an instrument that has a holistic approach, studies that show how EA impacts e-

Government performance are still in their infancy. Thus, this research begins by

developing a model to illustrate the relationship between EA and e-Government

system performance. This allows this study to respond to its focal question: to what

extent can EA be used to improve the Indonesian Treasury e-Government system

performance? The following questions were used to lead into this focal question:

a. Internationally, has EA implementation demonstrated benefits to e-

Government performance?

b. How can the quality or the process of EA development affect the e-


c. What is the current state of the Indonesian Treasury e-Government systems?

How can it be improved?

d. How can EA be used to improve the quality of the Indonesian Treasury e-

Government systems?

1.5 Methodology

In developing the framework, this study mainly referred to e-Government and

Enterprise Architecture theories. The study modified the EA Benefit Model (EABM)

from Tamm et al. (2011) to depict the correlation between the maturity of EA and the

e-Government system performance. The EABM was developed through a rigorous

literature review in the EA field. The quality of EA in the EABM is represented by the

maturity of EA and e-Government system performance represents the benefit gained

by the government. The research questions were answered by validating the research

framework using a single case study approach.

8

The case study is the Indonesian Treasury. In order to increase the quality of the

findings from the one case study, this study adopted a mixed method approach

(Garson, 2013a). The Indonesian Treasury’s employee views at various levels in the

organisational structure were captured. Therefore, relevant and valid data with low

bias was available for analysis.

The use of mixed methods in this study was also relevant to the nature of EA

development. It requires sound strategies and vision from high level officials (Cullen,

2012; Finkelstein, 2006; Gregor et al., 2007; National Information Society Agency,

2011). Thus, high level officials’ perspectives from Echelon three2 and above were

captured through semi-structured interviews.

Lankhorst (2004) argues that a top-down approach will most likely be lacking in

detail. Subsequently, Reithhofer and Naeger (1997) and Lankhorst (2004) argue that

in developing EA via the bottom-up approach should also be used to get a complete

picture. Therefore, this study gathered views at the operational level of the Indonesian

Treasury through quantitative research using questionnaires. As a result the study

collected adequate information to formulate a sustainable solution for the Indonesian

Treasury e-Government system transformation. Further details of the research

methodology are presented in Chapter 4.

1.6 Research significance

With a central role in managing public expenditure (Achmad, 2012), the

performance of the Indonesian Treasury e-Government has a significant impact on

Indonesian public services. Additionally, the Indonesian Treasury was one of the

2 Echelon three is upper level manager who is directly supervised by a director or head of a

Treasury Regional Office.

9

pioneers for reform initiatives in Indonesia (Nasution, 2003; The Indonesian Ministry

of Finance, 2011). Therefore, a study of the Indonesian Treasury e-Government system

could improve public services in Indonesia. Findings in this study could also be of

benefit to international donors who continuously support e-Government in developing

countries.

The future of e-Government research should be in relation to improving channels

to help government interact with its stakeholders (Löfstedt, 2005). Löfstedt (2005)

mostly refers to e-Government in developed countries in her study. Recent e-

Government studies followed this direction by using both a technological adoption

perspective (Carter & Bélanger, 2005; Kavanaugh et al., 2012; Reddick & Norris,

2013; Sharif, Davidson, & Troshani, 2013; Susanto, 2012; Susanto & Goodwin, 2013)

and an e-readiness perspective (Alghamdi et al., 2011, 2014; Koh et al., 2008).

However, very little research has examined the effect of holistic design

approaches, such as enterprise architecture (EA), on e-Government studies.

Developing countries are considered to have significant gaps between design and

implementation (Dada, 2006; Heeks, 2002, 2003) in their e-Government system

development. Hence, more work is needed in e-Government research from an EA

perspective to improve areas such as interoperability of data (Pardo, Nam, & Burke,

2012), cost efficiencies (Finkelstein, 2006; Lankhorst, 2009; Ross et al., 2006), and

business agilities (Op't Land et al., 2009; Saha, 2010b).

The number of research papers demonstrating the use of EA in e-Government is

growing (Bellman & Rausch, 2004; Chief Information Officer, 2005; Hald, 2006;

Mikaelian et al., 2011; National Information Society Agency, 2011; Østergaard

Jensen, 2010; Pardo et al., 2012; Pheng & Boon, 2007; Seppanen et al., 2009).

10

However, very few use empirical evidence that demonstrates the correlation between

the two. The aim of this study was to understand the effect of EA on e-Government in

developing countries. The study begins with the development of its research

framework by combining theories from both e-Government and EA.

This study uses data from the Indonesian Treasury as its empirical evidence. The

Indonesian Treasury has a role in both the front and back office in relation to

Indonesian public expenditure. According to Indonesian financial laws, the Indonesian

Treasury has a responsibility to authorise payment orders from spending units, book

keeping all revenue transactions, maintaining the asset management records and

generating both semester and annual government financial reports.

With the roles and responsibilities that the Indonesian Treasury has, its e-

Government system can be considered to be representative of the complexity in

analysing e-Government systems within the country, as Dawes (2008, p. 104) says:

Most information-intensive work actually takes place behind the scenes of public

web sites, in the “back offices” of government agencies. Here is where the

necessary policies and strategies are developed and the associated processes,

systems, and data resources are devised, managed, and used. In this work,

information technology is intertwined with public policy and management

concerns.

Consequently, this study is significant, because it:

1. Identifies the development of EA in developed countries’ governments. To

achieve this objective, this research provides in-depth analysis, mainly

based on secondary data from different governments. This then is used to

extrapolate the impact of EA on e-Government. The analysis also

demonstrates the importance of EA for other countries. Learning from other

countries’ experiences helps in understanding the value of EA. Thus, this

11

then gives stronger arguments about the importance of EA for the

governments of developing countries.

2. Investigates the relationship between EA and e-Government performance.

To do so, this research analyses currently available models to depict the

effect of EA on organisations in both public and private sectors. The study

modified the EA Benefit Model (EABM) to make it relevant in portraying

the relationship between EA and e-Government.

3. Assesses e-Government initiative in a major developing country. The

Indonesian Treasury is one major government institution that could have a

big impact on the Indonesian government. The result of this evaluation may

give a better view of how e-Government initiatives in developing countries

could be improved.

4. Discovers how EA can foster the Indonesian Treasury e-Government

system transformation. Extensive fieldwork is used to gather data and to find

out what is missing in the Treasury’s system transformation process and

how that process could be enhanced. The result of this could lead to an

understanding of how EA impacts e-Government in developing countries.

Thus, sustainable and continuous e-Government system improvement by

aligning business and technology could be facilitated.

1.7 Thesis outline

The outline of this thesis is illustrated in Figure 1.1 and described below.

12

Figure 1.1. The Structure of The Thesis

Chapter 1 introduces the research, the aims of this study and the research

questions. This chapter also defines the limitations of the study and briefly describes

the methodology to be used.

Chapter 2 reviews theories relevant to addressing the identified research

questions. Chapter 5 presents the Indonesian Treasury e-Government as the context of

this study. Chapter 3 develops the model to act as a research framework. The

Chapter One – Introduction

Literature Review

Chapter 5 – Research Methodology

Chapter 9 – Conclusions and Further Research

Chapter 2

E-Government, Enterprise Architecture

Chapter 3

The Indonesian Treasury e-Government Systems

Chapter 4

Research Model

Data Analysis

Chapter 6 Qualitative Analysis

Chapter 7 Quantitative Analysis

Chapter 8 – Discussions

13

framework is used to generate research instruments such as a list of questions for

interview and survey questionnaires.

Chapter 4 explains the methodology used in collecting the data and also

discusses the reasons for selecting the methods and techniques. Chapter 6 to Chapter

8 presents and discusses the findings of this study. Chapter 9 concludes the study and

outlines further research opportunities.

1.8 Conclusion and Summary

This study investigates the role of EA in improving e-Government performance

in developing countries. It focuses on the Indonesian Treasury e-Government systems.

The Indonesian Treasury is considered to be the core of the Indonesian public finance

systems. The system caters for revenue and expenditure data, national and regional

budgets and assets and government financial reports. Findings from this study may be

adopted by other government agencies in Indonesia and other developing countries.

Further, international donors, such as the World Bank may also benefit from this study.

Reviews of literature in the field of e-Government and EA are presented in Chapter 2.

14

LITERATURE REVIEW

2.1 Overview

This chapter reviews literature in the fields of e-Government and Enterprise

Architecture (EA). It begins with a literature review of the e-Government field in

Section 2.2. Section 2.3 reviews EA, an instrument that could be used to ensure

alignment between technology and business, and continuous development. Section 2.4

demonstrates the benefit of EA on e-Government by examining what developed

countries have achieved after adopting EA in their e-Government systems. The

presence of characteristics of EA in developing countries is presented in Section 2.5

and 2.6. This chapter is summarised in Section 2.7.

2.2 E-Government

Although implementation of ICT is meant to help improve the way governments

work (Ebrahim & Irani, 2005), governments are considered to be slow in adopting

proper technology to improve their public services (Marche & McNiven, 2003). The

use of ICT in supporting government work is called e-Government. E-Government

was first introduced in 1997 (Relyea, 2002). Gupta and Jana (2003) argue that e-

Government has become essential to improve national governance.

2.2.1 The evolution of e-Government

In order to give a sufficient view of e-Government evolution, it is important to

review how governments ran their business before ICT was used.

2.2.1.1 Antecedents to the evolution of e-Government

Government are considered to be complex enterprises (Fountain, 2001; Saha,

2010a). Governments also need to obey the laws and regulations in delivering its

15

public services, including providing information to the people (Fountain & Osorio-

Urzua, 2001). Furthermore, they are also typically more hesitant to change and to adopt

new initiatives than private companies (Marche and McNiven, 2003 in Davison et al.,

2005).

Weber introduced bureaucracy to deal with this complex role of government

(Fountain, 2001, pp. 44-45). Bureaucracy, referring to government officials who run

the public administration, is often interpreted as complicated with inefficient processes

resulting in dysfunctional division and lack of responsiveness (Duivenboden and Lips,

2005 in Homburg, 2008).

The nature of Weberian bureaucracies is very systematic, based on rules, and

with hierarchical authorisation systems designed to implement government policies

and programs (Rainey, 2009; Welch & Pandey, 2007). Paper is used as the medium of

government work. The paper is sent back and forth to gain approvals according to the

hierarchy in the government organisation (Hughes, 2003). This complexity in ensuring

accountability and responsiveness results in bureaucracy being synonymous with red

tape (Welch & Pandey, 2007). Furthermore, bureaucracy causes negative impacts to

government administration such as corruption (Hillman, 2004; Rauch & Evans, 2000;

Van Rijckeghem & Weder, 2001) and incongruity between bureaucracy and society

(Haque, 1997).

In order to overcome performance related issues in delivering public services,

governments reformed their public management (Keating, 1998). Implementation of

New Public Management (NPM) became common including in Asian governments

(Samaratunge et al., 2008; Schick, 1998). The main objective of the NPM was to

16

increase efficiency and effectiveness of organizations in the public sector (Keating,

1998).

The objectives of NPM are considered to be a fit with the use of ICT in

government (Homburg, 2004; Torres et al., 2005). If used properly, a robust ICT

investment (Irani, 2002) may be able to simultaneously enhance effectiveness (Carter

& Bélanger, 2005; Pardo et al., 2012; Shim & Eom, 2008), sustain accountability

(Welch & Pandey, 2007), transparency (Ciborra, 2005), and eradicate corruption (Cho

& Choi, 2004; Shim & Eom, 2008).

2.2.1.2 Theoretical bases for the Evolution of e-Government

The use of ICT in government can be traced back to the 1970s (Danziger &

Andersen, 2002). They define ICT as instruments that are used to digitally manage

data and information. Adopting ICT in government should not simply be defined as

being online (Curtin et al., 2003). As a system, e-Government involves various entities

such as technological, socio-political, and governmental (Homburg, 2008). As such e-

Government development should not be seen as an easy process.

E-Government is used to modernise public administration (Lenk, 2002). Kofi

Annan (2001, p. 1), former UN Secretary General, stated that:

“ICT is not a magic formula that is going to solve all our problems. But it is a

powerful tool with diverse applications. Our challenge is to put that power at

the service of all humankind”.

By optimising the use of ICTs, the OECD believes that governments will be able to

enhance their organisational structures, focus further on citizens’ needs and improve

their performance (OECD, 2003).

17

2.2.2 Definition of e-Government

E-Government is considered to be a relatively new field of research (Curtin et

al., 2003; Grönlund, 2008; Reece, 2006). To date there is no single, universal definition

of e-Government. E-Government is not simply defined as providing information

through a website on the Internet (Curtin et al., 2003). This limited view of the Internet

being the main, or sole, indicator for the presence of e-Government may restrict what

is possible with ICT in government (Yildiz, 2007).

To date there are several definitions of e-Government. The World Bank (2011,

p. 1) defines e-Government as:

the use by government agencies of information technologies (such as Wide Area

Networks, the Internet, and mobile computing) that have the ability to transform

relations with citizens, businesses, and other arms of government.

According to Siskos, Askounis, and Psarras (2014, p. 1), e-Government refers to:

the use of information and communication technologies (ICT) by governments

to provide digital services to citizens and businesses over the Internet, at local,

national or international level.

Another definition of e-Government comes from Guo (2010, p. 1):

a way for governments to use the most innovative information and

communication technologies, particularly web-based Internet applications, to

provide citizens and businesses with more convenient access to government

information and service.

Joon (2009, p. 2) defines e-Government as:

the government’s efforts to transform both internal and external governmental

relationships through the use of information technology such as the Internet.

The OECD (2003) refers to the following e-Government definition:

the use of Information and Communication Technologies (ICT) and,

particularly, the Internet as a tool to achieve better government

18

Before coming to a general definition, the OECD (2003, p. 23) classifies e-

Government into three definitions:

1. e-Government is defined as Internet (online) service delivery and other

Internet-based activity such as e-consultation.

2. e-Government is equated to the use of ICT in government. While the focus

is generally on the delivery of services and processing, the broadest

definition encompasses all aspects of government activity.

3. e-Government is defined as a capacity to transform public

administration through the use of ICT or indeed is used to describe a new

form of government built around ICT. This aspect is usually linked to

Internet use.

Since the aim of this study is to investigate how well the Indonesian Treasury e-

Government system performs, this study makes use of the second definition of e-

Government from the OECD.

2.2.3 Implication of e-Government

The implementation of e-Government can affect public organisations’ business

processes, technologies and human resources (Silcock, 2001). Therefore governments

need to analyse advantages and disadvantages before developing and deploying e-

Government systems (Bock, 2005). Thus, the government should be prepared for any

unforeseen surprises as a result of e-Government implementation (Reece, 2006).

e-Government is critical in transforming public administration (Finger &

Pécoud, 2003; Torres et al., 2005; Ebrahim & Irani, 2005), to provide convenient and

accessible public services (Carter & Bélanger, 2005), to improve accountability

(Carter & Bélanger, 2005; Gil-Garcia & Martinez-Moyano, 2007; Wong & Welch,

2004) and to enhance transparency (Bertot et al., 2010; Ciborra, 2005; Jaeger & Bertot,

2010; Kim et al., 2009; McDermott, 2010). To date, then ICT improvement has forced

government to adjust their technology usage and/or business functions (Dener et al.,

2010).

19

The implementation of e-Government would therefore change how government

operates (Ebrahim & Irani, 2005). Tat-Kei Ho (2002) argues that e-Government

implementation changes government perspectives in many ways, such as shifting from

bureaucratic views to a stakeholder oriented mindset. He further mentions that there

are eight criteria that would change, i.e., orientation, process organisation, leadership

style, internal communication, external communication, mode of service delivery, and

principles of service deliveries as seen in Table 2.1.

Table 2.1. Shifting from bureaucracy to e-Government (Tat-Kei Ho, 2002)

Criteria Bureaucratic e-Government

Orientation Production cost-efficiency User satisfaction and control, flexibility

Process organization Functional rationality, departmentalization, vertical hierarchy of control

Horizontal hierarchy, network organization, information sharing

Management principle

Management by rule and mandate

Flexible management, interdepartmental team work with central coordination

Leadership style Command and control Facilitation and coordination, innovative entrepreneurship

Internal communication

Top-down, hierarchical Multidirectional network with central coordination, direct communication

External communication

Centralized, formal, limited channels

Formal and informal, direct and fast feedback, multiple channels

Mode of service delivery

Documentary mode, and interpersonal interaction

Electronic exchange, non-face-to-face interaction

Principles of service delivery

Standardization, impartiality, equity

User customization, personalization

20

The changes are considered to be apparent in developed countries with more

users satisfied with e-Government deliverables (Reddick & Roy, 2012; Verdegem &

Verleye, 2009). However, governments in developing countries are still struggling in

making the most of their e-Government initiatives (Alghamdi et al., 2014; Dada, 2006;

Heeks, 2003; Rokhman, 2011).

2.2.4 e-Government in Developing Countries

E-Government has not yet reached its potential in developing countries. Several

reasons for this are that governments have not been keen to redefine their business and

remain slow in adopting proper technology to support their objectives (Ndou, 2004).

This situation leads to complications in e-Government implementation. Despite these

complications, governments are keeping their e-Government projects running

(Sandeep & Ravishankar, 2014).

The risk in developing e-Government implementation remains high (Ndou,

2004). Therefore, adopting established studies for the area of e-Government research

is needed. The technology adoption perspective has received significant attention for

e-Government research in developing countries (Alzahrani, 2014; Carter & Bélanger,

2005; Mirchandani et al., 2008; Susanto & Goodwin, 2013).

2.2.5 Technology Adoption Perspective of e-Government

2.2.5.1 Technology Adoption Model (TAM)

The Technology Adoption Model (TAM) is considered to be the most popular

model used to validate e-Government studies from a technology adoption perspective

(Susanto, 2012). Within a decade from when it was introduced in 1989, TAM had

become a robust model to predict user acceptance of technology (Venkatesh & Davis,

2000). TAM was built on the Theory of Reasoned Action (TRA) (Davis et al., 1989).

21

TAM, illustrated in Figure 2.1, proposes external variables such as training, anxiety,

computer related support and experience, all affect cognitive responses that manifest

as perceived usefulness and perceieved ease of use. Both cognitive responses may

influence attitude toward using. Attitude toward using together with perceived

usefulness may affect behavioural intention before, finally, being reflected in actual

behaviour.

Figure 2.1. Technology Acceptance Model (TAM) (Davis et al., 1989, p. 985)

TAM has been used extensively in e-Government studies. Mirchandani et al.

(2008) used TAM to conduct multigroup analysis of citizen perspectives toward e-

Government in Thailand and Indonesia. TAM was also used to evaluate e-Government

acceptance in Macao (Lai & Pires, 2010). Susanto and Goodwin (2013) developed the

SMS-based E-Government Acceptance Model (SEGAM) by extending TAM to study

user acceptance of SMS-based e-Government services.

TAM has been extended to TAM2 by imposing moderators from two external

variables: Experience and Voluntariness (Venkatesh & Davis, 2000). TAM2 argues

that cognitive responses may be moderated by influence from other people.

2.2.5.2 The Unified Theory of Acceptance and Use of Technology (UTAUT)

Venkatesh et al. (2003) integrated eight prominent theories including TAM and

TRA into their model called the Unified Theory of Acceptance and Use of Technology

ExternalVariables

PerceivedUsefulness (PU)

PerceivedEase of Use (PEU)

AttitudeToward Using (A)

Behavioural Intention to Use

(BI)

Actual System Use

22

(UTAUT). It predicts user’s acceptance of new technology (Venkatesh et al., 2003).

Moreover, the UTAUT assists understanding the acceptance factors of related

activities such as training to adopt and implement new systems. The UTAUT is

depicted in Figure 2.2.

Figure 2.2. The Unified Theory of Acceptance and Use of Technology (UTAUT)

(Venkatesh et al., 2003, p. 447)

Similar to TAM, UTAUT gained significant attention in e-Government studies.

For instance Verdegem and Verleye (2009) built a model on UTAUT to measure user

satisfaction in e-Government. A case study of citzen adoption of web-based e-

Government in Saudi Arabia also used UTAUT (Alzahrani, 2014). UTAUT has also

been used to develop a model determining factors influencing intention to use e-

Government in Malaysia (Lean et al., 2009).

Consequently, both TAM and UTAUT have been widely used to predict users’

acceptance toward newly adopted technology in various countries. To evaluate

technology adoption from the organisation perspective, the Technology-Organisation-

Environment (TOE) framework is an option.

PerformanceExpectancy

EffortExpectancy

Facilitating Conditions

PerformanceExpectancy

Social Influence

Gender Age ExperienceVoluntariness

of Use

Use Behaviour

23

2.2.5.3 The Technology–Organisation–Environment (TOE)

The TOE framework was introduced to understand factors that lead to

technology adoption from the organisational perspective (Depietro et al., 1990).

Depietro et al. (1990) illustrate that technology adoption is influenced by three factors,

namely: technology, external task environment and organisation as seen in Figure 2.3.

Figure 2.3. The Technology-Organisation-Environment (TOE) (Baker, 2012, p.

236; Depietro et al., 1990, p. 153)

The TOE framework has been used in e-Government studies. Sharif et al. (2013)

investigated social media adoption in local government agencies by using the TOE

framework. The collective impact of e-Government and e-Business on Singapore’s

economic performance was empirically validated using the TOE (Srivastava & Teo,

2010). The TOE has also been used to validate determinants of e-Government

integration in developing countries such as Indonesia (Pudjianto et al., 2011).

2.2.5.4 Limitations of Technology Adoption in Developing Countries

While technology adoption has received significant attention from researchers

in e-Government studies, studies investigating how e-Government can be managed

External Task Environment

Industry Characteristics and Market Structure

Technology Support Infrastructure

Government Regulation

Organisation

Formal and Informal Linking Structures

Communication Processes

Size

Slack

Technology

Availability

Characteristics

Technological Innovation Decision Making

24

from a more holistic enterprise architecture perspective in developing countries remain

limited. Such a study requires sufficient access to sources of information from

government officials and regulations. Accessing sources of information that are not

publicly available in developing countries is considered difficult (Banisar, 2006;

Poulin, 2004).

As a complex enterprise, government requires an instrument to ensure the

alignment of business and technology in e-Government initiatives. Such an instrument

should cover all aspects not only within its government agencies but also in the relation

with its stakeholders (Saha, 2010b). Business-ICT alignment may be achieved by

adopting EA (Gregor et al., 2007; Lankhorst, 2009; Liimatainen, 2008).

This study uses EA as the means for improving e-Government alignment in

developing countries. Improved alignment should enhance understanding of e-

Government in a broader sense (Walsham & Sahay, 2006). Moreover, it is evident that

developing countries still have limitations, such as infrastructure and human resources,

which could impact the quality of e-Government systems (Walsham & Sahay, 2006).

Therefore, in implementing e-Government systems, governments of developing

countries require holistic approaches relevant to their contexts (Ndou, 2004; Walsham

& Sahay, 2006) to address transparency, accountability and accessibility (Sandeep &

Ravishankar, 2014).

2.2.6 Developing and Deploying e-Government

Implementing e-Government needs a proper strategy. The strategy should refer

to internationally acknowledged formulation (Basu, 2004). To achieve the objective

of e-Government, the formulated strategy requires strong commitment from all parties

who are involved in its development and deployment (Basu, 2004). The e-Government

25

implementation strategy also requires strong integration across the unique functions of

multiple levels of the government (Layne & Lee, 2001).

Commonly, e-Government research is about increasing citizen involvement

(Carter & Bélanger, 2005; Cook, 2000; West, 2004); improving public accessibility to

government services such as through SMS (Lallana, 2004; Rossel et al., 2006; Susanto,

2012; Trimi & Sheng, 2008), or Internet websites (Ahn, 2012; Gil-Garcia, 2005;

Jaeger, 2004; Wong & Welch, 2004). Although Layne and Lee (2001, p. 125) say:

the critical benefits of implementing e-Government are actually derived from the

integration of underlying processes not only across different levels of

government but also different functions of government.

However, few studies have been undertaken in the area of integration and

interoperability among government units. These areas are considered to be within the

domain of EA (Saha, 2009, 2010b). Therefore, the presence of EA could be used to

assist in achieving the strategic values of e-Government.

2.2.7 Achieving Strategic Values of e-Government

One of the challenges is that most governments consider having an appropriate

strategy for implementing e-Government to be of low importance (Aichholzer, 2004;

Shahkooh & Abdollahi, 2007). In addition to successful e-Government

implementation, having an appropriate strategy also leads to sustainable development

in the long run.

In order to have sustainable e-Government development, government should

have a clear vision of what will be delivered. Such vision should be articulated clearly

in relevant documents (Grant & Chau, 2006). Such documents enable the development

process to be executed properly by relevant government agencies. This will result in

measurable performance benefits of e-Government systems.

26

2.2.8 Measuring e-Government Performance

This study defines e-Government performance as the quality of the services it

delivers. There are several models available to measure e-Government performance.

Peters et al. (2004) classify them into three groups: stage models, service quality, and

performance indicators.

2.2.8.1 Stage Models

Stage models identify the maturity of e-Government through how governments

exploit ICT, such as the Internet, to interact with their stakeholders including citizens,

private sectors, and other government bodies. Although most of the models in this

group start with the static government website, the path by which each of the models

progresses is different. Most of the models start with the official presence of an e-

Government website characterized by simple information distribution (Hiller &

Bélanger, 2001; Layne & Lee, 2001; Moon, 2002; Shahkooh et al., 2008; Siau & Long,

2005; Wescott, 2001). Table 2.2 shows the evolution of e-Government stages as

described by the authors (in alphabetical order) referred to above.

From this Table it can be seen that most of the models view e-Government by

how governments utilize it to improve public participation (Bellamy & Taylor, 1994;

Danziger & Andersen, 2002; De Araujo, 2001; Edelenbos & Klijn, 2006; Homburg,

2008; Torres et al., 2005). Therefore, these models can be used to measure e-

Government where the Internet is used extensively. These models are, however,

lacking empirical evidence. For that reason, Coursey and Norris (2008) conducted an

empirical study to find out whether these models can be used to ensure that e-

Government is progressing as predicted by the model. They found that it cannot be

clearly perceived that e-Government is progressing in accordance with the proposed

stages.

27

Table 2.2. E-Government Stages

Author Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Stage 6

Hiller and Bélanger (2001)

Information dissemination

Two-way communication

Integration Transaction Participation

Layne and Lee (2001)

Catalogue Transaction Vertical Integration Horizontal Integration

Moon (2002) Information dissemination


Service and financial transaction

Vertical and Horizontal Integration

Political participation

Shahkooh et al. (2008)

Online presence Interaction Transaction Transformation Digital Democracy

Siau and Long (2005)

Web presence Interaction Transaction Transformation e-Democracy

The United Nations (2003)

Emerging presence Enhanced presence Interactive presence

Transactional presence

Network presence

Wescott (2001) e-mail and internal network

Internal organization & public access to information


Exchange of value Digital democracy Joined-up government

28

The United Nations has been adopting the stage model to measure worldwide e-

Government readiness (The United Nations, 2003, 2004, 2005, 2008, 2010, 2012). The

stage model may be relevant to recognize at what stage the e-Government is, but

relying solely on one of the models may lead to insufficient interpretation of e-

Government performance because the quality of public service delivery is excluded in

the model.

2.2.8.2 Service Quality

The main role of a government is delivering services to the people (De Araujo,

2001; Kellough & Nigro, 2006). The characteristics of services are intangible

(Parasuraman et al., 1985), it requires an instrument that can measure the effectiveness

of how a service is being delivered, e.g., service quality. Service quality relies on

satisfaction and perceived quality (Peters et al., 2004). It was Parasuraman et al. (1988)

who designed an instrument called SERVQUAL to measure perceptions of service

quality. Parasuraman et al. (1988) argue that customers’ expectations and perception

of quality can be used to measure service quality. In order to do so, they developed a

questionnaire with 22-item grouped into five dimensions, namely: tangibles,

reliability, responsiveness, assurance, and empathy (Parasuraman et al., 1991;

Parasuraman et al., 1988).

Parasuraman et al. (1988) categorise physical facilities, equipment and personnel

as part of the tangibles dimension. They define the reliability dimension as accuracy

and the ability to deliver the promised service. Willingness to give immediate support

is put into the responsiveness dimension. As for the assurance dimension, they include

sufficient knowledge to stimulate trust and confidence. In the last dimension, empathy,

understanding individual customer profiles is used.

29

Initially, SERVQUAL was proposed in marketing studies. Information Systems

(IS) studies adopted the model to measure IS effectiveness (Gorla, 2012; Kaisara &

Pather, 2011; Kettinger & Lee, 1994; Myerscough, 2002; Petter et al., 2008; Pitt et al.,

1995; Van Dyke et al., 1997; Watson et al., 1998). Although the five dimensions in

SERVQUAL have been stable over time, these dimensions cannot be considered

universal for all disciplines (Bigne et al., 2003). Therefore the use of SERVQUAL in

IS research often requires modification (Bigne et al., 2003; Gorla, 2012).

Gorla (2012) modified the instrument to measure IS success by restructuring the

questions. This modified instrument has become prominent because it can measure

comprehensive ICT service, rather than individual products, delivered by an

organization. Thus, it is able to measure whether or not the expected benefits have

been achieved.

2.2.8.3 Performance Indicators

In determining the performance of government websites, Lee (2003) proposed

measurements based on modifications of the Simeon (1999) studies. His proposed

measurement components are as follow:

1. Attracting. It is meant to impress the users and is determined by several items,

e.g., the logo and tagline; graphics (including the choice of colours, layout, and

themes); self-advertisement from the website’s owner; attracting services such

as quizzes and maps; features for attracting such as a gallery, tourism, game,

and finance.

2. Informing. It consists of modified Simeon (1999) components, e.g., local links,

content for publicity, content for learning, reports, descriptions of the

30

institution, description of the online administrative services, projects, contact

information and counseling.

3. Community comprised of ten items: online forum, events, partner links,

newsletter, message boards, user participation, focus of news, vision or values,

domain identity and community services.

4. Delivering is a binary variable to identify the presence or absence of features

for each essential item such as a search engine, mailing list, framework,

multimedia, password system, downloadable publications.

5. Innovation is provided to measure whether or not public institutions have

utilized the Internet with sufficient service innovation.

Although Peters et al. (2004) put this two-stage model for monitoring website

strategy in the performance indicators group, this model only measures what is visible

on the interface of a web site. This group is viable in assessing what governments offer

on their Internet website. Yet this model may not be suitable to determine how

government is performing in delivering their services based on the utilization of ICT

in alignment with how governments conduct their businesses.

2.3 Enterprise Architecture

A sound alignment between ICT and business will definitely deliver benefits to

any organization that is able to put it into action (Gregor et al., 2007; Pereira & Sousa,

2005). As a result, outcomes will be improved. Strategic alignment theory defines

alignment as a never ending and continuous improvement through understandable

performance indicators, framework, good governance and harmonious process

execution (Venkatraman et al., 1993). In constructing this alignment, organizations

have made ICT their main tool in delivering business objectives (Gregor et al., 2007).

31

The fundamental alignment should clearly be defined in their business and ICT

strategic planning.

In order to achieve an alignment of ICT and business that includes resources

such as human capital, organization and knowledge, it is important to have

comprehensive management and governance instruments. One tool that can be used to

generate this alignment is Enterprise Architecture (Elhari & Bounabat, 2011;

Finkelstein, 2006; Gregor et al., 2007; Martin, 2005; Pereira & Sousa, 2005; Ross et

al., 2006).

2.3.1 The Evolution of Enterprise Architecture

In 1997, Zachman, after working on information systems architecture for a

decade, introduced the terminology of Enterprise Architecture to the Information,

Communication and Technology (ICT) industry.

2.3.1.1 Antecedents to the evolution of Enterprise Architecture

To date, “change” still remains the biggest challenge that modern enterprises

need to deal with (Zachman, 1997). He argues:

there are only three options for managing enterprise change: by trial and error;

by reverse engineering; or by going out of business! (Finkelstein, 2006, p. 9)

Hence, proper plans and strategies are needed to perform necessary changes. In the

information systems related industries, shifting from strategy to implementation

remains an issue (Zachman, 1997). Other industries, such as the airplane or

manufacturing industries, have fewer issues in shifting from strategies to

implementations. Architecture was one of the reasons for why this is possible

(Zachman, 1997).

32

Zachman (1997) further argues that the complexity of development and

implementation processes in information systems is similar to the product

development processes in the airplane and manufacturing industries. Therefore, he

suggests an architecture that could be used to not only manage the complexity of ICT

implementation from its strategy but also to capture alignment of relevant aspects in

information systems development. It is called Enterprise Architecture.

2.3.1.2 Theoretical bases for the Evolution of Enterprise Architecture

Bernard (2012) argues that the development of EA was influenced by

organisational theory and systems theory. “Enterprise architecture captures the

essentials of the business, IT and its evolution. The most important characteristic of an

enterprise architecture is that it provides a holistic view of the enterprise” (Lankhorst,

2009, p. 3).

Zachman (1987) was the first to introduce the EA framework, which was initially

called the information system architecture. After that he continuously improved it by

extending and formalizing the framework (Zachman & Sowa, 1992). The final result

was the Zachman Framework for Enterprise Architecture (Zachman, 1997). Zachman

(2003) developed this framework in order to achieve the reusability principle. The idea

was to try to replicate practices from the construction and aircraft industries such as

doors that can be reused for different targeted objects (Zachman, 2003).

The Zachman framework for Enterprise Architecture is represented by a two

dimensional matrix of five rows and six columns as can be seen in Figure 2.4. The last

row is generally used to show the functions of an enterprise. In the framework, the six

columns represent the objectives. They contain basic questions (what, how, where,

who, when and why) that require a specific answer. Then, the five rows list the subjects

33

or roles that are involved in the process, i.e. planner, owner, designer, builder, and

subcontractor. The intersection of a column and a row is called a cell.

Figure 2.4. The complete Zachman Framework for Enterprise Architecture

(Zachman, 2003 cited in Finkelstein, 2006, p. 4)

Unlike the traditional system development that uses a bottom-up view, the

Zachman framework is a top-down approach (Finkelstein, 2006). In the traditional

model the system development process begins by looking at the current manuals or

existing systems. The findings are then analysed to improve them. Normally, the

requirements are defined by ICT people through extensive interviews to understand

the business needs. This is likely to result in a design based on technology. It will then

become a technology dependent solution. As a result, changes to the business will be

difficult to resolve with the ICT.

The first three rows of the Zachman Framework are technology independent.

The business people create the strategic direction for the future business as a result of

these three rows. Therefore, Finkelstein (2006) states that the output of the strategic

direction will be in business terms. Hence, in order to develop systems that meet future

34

business needs, early involvement by the ICT department is essential (Finkelstein,

2006). With a clear strategic plan, Finkelstein (2006) further states that the ICT

department will be able to produce technology/system requirements comprised of the

designs for the application, database and technology. Once the system is implemented,

the performance can be measured through predefined performance requirements.

These are the results of rows three to five in the framework.

In addition to the Zachman Framework for Enterprise Architecture, Spewak and

Hill (1993) introduced Enterprise Architecture Planning (EAP). EAP is used to define

a planning process that includes business strategy, process reengineering,

standardization, and system design (CIO Council, 1999; Janssen & Hjort-Madsen,

2007; Minoli, 2008; Rohloff, 2008).

In 2001, the Open Group introduced another industry standard architecture

framework, namely TOGAF. In the first release, TOGAF version 7 focused on

technical architecture. Two years later, the open group introduced TOGAF version 8.1

to cater for the entire enterprise scope, i.e., business, information, application and

technical architecture (Saha, 2004). The latest version of TOGAF, version 9.1, was

introduced in 2009. Some changes were made to the previous version to focus more

on holistic enterprise change (The Open Group, 2009). Since TOGAF was introduced

later, only the Zachman Framework for Enterprise Architecture and Enterprise

Architecture Planning were used by the United States to develop the Federal Enterprise

Architecture Framework (FEAF).

2.3.2 Definition of Enterprise Architecture

EA is considered to be a relatively new field of study in information systems

(Harmon, 2003; Langenberg & Wegmann, 2004; Lankhorst, 2009). Initially, EA was

35

called an information systems architecture (Zachman, 1987; Zachman & Sowa, 1992).

In 1997, Zachman started using the terminology of Enterprise Architecture.

Although EA has existed for over a decade (Zachman & Sowa, 1992), to date

there is no single definition for EA (Janssen & Kuk, 2006; Stelzer, 2010). The first

definition of EA came from Zachman (1997, p. 6) who defines it as:

set of descriptive representations (i.e. ‘models’) that are relevant for describing

an Enterprise that can be produced to management’s requirements (quality) and

maintained over the period of its useful life (change).

Langenberg and Wegmann (2004, p. 2) define EA as:

blueprint that documents all the information systems within the enterprise, their

relationships, and how they interact to fulfil the enterprises mission.

Ross et al. (2006, p. 9) define EA as:

the organizing logic for an organization’s core business processes and ICT

capabilities captured in a set of policies and technical choices, to achieve

business standardization and integration requirements of the firm’s operating

model.

The Open Group Architecture Framework (TOGAF) (2009, p. 5) define it as:

any collection of organizations that has a common set of goals encompassing all

of its information and technology services, processes, and infrastructure.

A recent definition of EA came from Saha (2010, p. 6), who defines it as:

the inherent design and management approach guided by principles,

frameworks, methodologies, requirements, tools, reference models, and

standards for organizational coherence leading to alignment, agility and

assurance.

The above definitions imply that EA applies beyond just technology. It represents the

alignment of business and technology.

36

2.3.3 Aligning Business and Technology by using Enterprise Architecture

ICT and business alignment consistently ranks in the top five concerns of high

level officials who rely on ICT in running their business (Chan & Reich, 2007;

Luftman, 2004). EA has the capability for realising the alignment of business and ICT

in a common framework (Gregor et al., 2007). It is evident that EA could enable

alignment of ICT and business in the public sector (Martin, 2005). Thus, EA is seen

as attractive with the governments of some developed countries adopting it as part of

their strategies (AGIMO, 2007, 2009; Chief Information Officer, 2005; CIO Council,

1999; Hanafiah & Goodwin, 2011; Joon, 2009; National Information Society Agency,

2011; Saha, 2009, 2010a, 2010b; U.S. Congress, 1996).

In adopting EA as part of government strategy, Martin (2005) argues that a

government agency should have detailed information on the expected business and

ICT processes, including the required strategies and resources based on its uniqueness.

Although the use of EA has become prominent, government needs to make sure that

EA adoption will be beneficial in addition to the other commonly adopted instruments.

2.3.4 The Use of Enterprise Architecture and Other Instruments

For years before some governments decided to adopt EA to boost their

performance, they had already adopted common standards. Lankhorst (2009) classifies

the standards into two main areas: general management and ICT management as seen

in Figure 2.5. In the area of general management, such as in strategic management,

most governments adopted balanced scorecard (see Griffiths, 2003; Lilian Chan, 2004;

Niven, 2011).

37

Strategic Management (Balanced Scorecards)

Strategy Execution (EFQM)

IT Governance (COBIT)

IT Delivery and Support (ITIL) Quality

Management (ISO 9000)

IT Implementation (CMMI)

General Management IT Management

Figure 2.5. Management areas related to Enterprise Architecture (Lankhorst,

2009, p. 14)

Some governments adopted the European Foundation for Quality Management

(EFQM) to execute the strategy (Jacobs & Suckling, 2007). The use of The

International Organization for Standardization (ISO) 9000 to ensure government

achieves desired quality (Bayo-Moriones et al., 2011; Jaap van den et al., 2005) is also

common.

In relation to ICT adoption, several instruments were introduced to achieve as

much as possible from the ICT investment. Control Objectives for Information and

Related Technology (COBIT), a framework to establish appropriate governance in

ICT, is also generally accepted by government (Gerke & Ridley, 2006; Ridley et al.,

2004).

The United Kingdom government developed the Information Technology

Infrastructure Library (ITIL) for service management purposes (Arraj, 2010). The ITIL

is adopted by the private and public sectors (for example see Cater-Steel & Tan, 2005;

Iden & Langeland, 2010; Pollard & Cater-Steel, 2009).

Carnegie Mellon University developed the Capability Maturity Model

Integration (CMMI) to measure ICT implementation, especially in software

development, and is another common standard adopted by governments (Mongkolnam

et al., 2009; Wu et al., 2006).

38

The previously mentioned standards remain important even with the adoption of

EA (Lankhorst, 2009). Lankhorst (2009) further argues that the presence of EA is

needed for general management within an organisation in order to:

1. enrich the Internal Business Processes of the Balanced Scorecards;

2. create a clear view and to communicate the Policy and Strategy and the

Processes features of the EFQM; and

3. integrate ICT support systems with the business processes as to how to

design, manage and document them.

EA can also be used in conjunction with other established IT management

standards such as COBIT, ITIL and CMMI (Lankhorst, 2009; Wilkinson, 2006).

Lankhorst (2009) argues that EA could extend such standards by:

1. creating a base line for the internal control governance of COBIT;

2. creating good references that would show dependencies among ICT

entities that would boost service management when using ITIL; and

3. providing guidelines and restrictions to assess software development

using CMMI.

In light of the above, the presence of EA would not interfere with the adoption

of common standards. This study argues that the presence of other established

standards could hasten the development of EA because organisations are already

familiar with internationally accepted frameworks.

2.3.5 Measuring the Progress of Enterprise Architecture

The progress of EA can be identified by measuring its maturity. To date, limited

maturity models have been proposed. This study found a list of EA maturity models

as seen in Table 2.3. Some models were developed by governments, e.g., Hite (2003),

39

the Office of Management and Budget (2005), the National Association of State Chief

Information Officers (NASCIO) (2003), and some models developed by prominent

consultants such as Gartner (Burke, 2010), EA researchers such as Ross et al. (2006)

and the Institute For Enterprise Architecture Development (IFEAD, 2004).

Table 2.3. EA maturity models

Author Year Publication title

Brian Burke (Gartner) 2010 ITScore for Enterprise Architecture

NASCIO 2003 Enterprise Architecture Model

Randolph C. Hite (the US

Government Accounting

Office/GAO)

2003 The Enterprise Architecture

Management Maturity Framework

(EAMMF)

Ross et.al. 2006 Enterprise Architecture as strategy

The IFEAD 2004 Extended Enterprise Architecture

Maturity Model (E2AMM)

The US Office of Management and

Budget (OMB)

2005 Enterprise Architecture Assessment

Framework

This study found that each model has strengths and weaknesses. A further

analysis of EA maturity is presented later in Section 3.5. This study proposes an

alternative model in the same chapter. The alternative model can be used to determine

whether or not EA makes a significant contribution to e-Government development.

2.4 Enterprise Architecture Contribution to e-Government in Developed Countries

Governments have long searched for ways to improve their public services. The

application of e-Government has made this objective achievable. E-Government

promises to modernize public services and enable governments to act better, faster,

smarter and more reliably to meet the needs of people. To achieve these anticipated

benefits, Information, Communication and Technology (ICT) needs to be properly

40

aligned to government business strategy. Recent studies of e-Government reveal that

there is an ICT planning gap in the transformation process (Davison et al., 2005).

Current research in information systems has acknowledged Enterprise

Architecture (EA) as an instrument to make integration of business and technology

viable. Ross et al. (2006) propose EA to be used in this alignment strategy. Zachman

(1987), the first designer of an EA framework, claims that EA could be used to manage

change and complexity in the enterprise. EA can also be used to promote

interoperability (Finkelstein, 2006; Martin, 2005) among and within enterprises such

as government agencies.

The complexity of public sector systems (Saha, 2010a) makes EA an attractive

option for governments. Governments in developed countries have introduced EA as

a key component of their holistic and coherent e-Government approaches. Although

the number of developed countries adopting EA is growing (Liimatainen et al., 2007),

very few studies have proven that EA leads to better e-Government performance.

From the United Nations (UN) e-Government development index 2003,

Shekkerman (2004) identified a strong correlation between rankings and EA activities.

However, Shekkerman (2004) relied on only one year of data. Replicating what he did,

the study used serial data from 2003-2012 from the UN e-Government development

index to investigate correlation between e-Government and EA programs in four

developed countries.

2.4.1 E-Government Development Index

The United Nations regularly produces an e-Government Development Index.

The first index was launched in 2003. In measuring the e-Government development

index, the UN adopted the stage model (Peters et al., 2004; K. A. Shahkooh et al.,

41

2008; Siau & Long, 2005). Stage models identify the maturity of e-Government

through how governments exploit ICT such as the Internet, to interact with their

stakeholders, e.g., citizens, private sectors, and other government bodies.

Research by Coursey and Norris (2008) found that a stage model cannot clearly

perceive that e-Government is progressing based on the stages. Instead, a stage model

may be relevant in recognizing at what stage the e-Government stands. In evaluating

how governments utilize ICT the UN used three major variables in their model.

2.4.1.1 Variables Measured

The three major variables used by the UN in their model, measure: 1) how well

the application of ICT is used to improve public services; 2) supporting infrastructure;

and, 3) human capital. The maximum result for each variable is one. All variables are

weighted equally. The aggregate of these weighted variables leads to the total e-

Government index for each country.

The application of ICT in UN member countries was measured by a web measure

index. A few years later, this variable was renamed to the online services index. This

variable was measured purely by a quantitative online survey on a particular

government website or portal on the Internet. The other two variables, supporting

infrastructure and human capital, refer to secondary data. Data for supporting

infrastructure were predominantly taken from the UN International

Telecommunication Union (ITU) and the UN Statistics Division. This variable is

referred to as the telecommunication infrastructure index. In generating the human

capital index, the UN depends on the United Nations Educational, Scientific and

Cultural Organization (UNESCO) data. Since the two variables rely on secondary data,

42

the dynamic in measuring the government index was partly changed to the application

used to improve public services.

2.4.1.2 Changes in the Measurements

Since it was introduced in 2003, the UN has continuously refined its model and

instruments. Initially, the objective was intended to measure the use of ICT,

particularly the Internet to deliver public services (The United Nations, 2003, 2004,

2005). After that, the view of e-Government as a whole concept was initiated in 2008

(The United Nations, 2008). As the number of mobile device users has grown rapidly

(The International Telecommunication Union, 2011), since 2010 the UN has included

mobile services (The United Nations, 2010). Consequently, the web measure index

was changed to the online services index. In addition to those changes, in the latest

publication, the UN measured inter-linkages that will lead to sustainability of e-

Government systems (The United Nations, 2012).

Evidence of the changes in the objective of the UN e-Government index can be

seen in the modification of the employed stage model. Originally, the UN stage model

comprised five stages, namely: Emerging Presence, Enhanced Presence, Interactive

Presence, Transactional Presence and Networked Presence. In 2008, the UN modified

the stages to Emerging, Enhanced, Interactive, Transactional, and Connected. The

modified stages were then simplified to Emerging, Enhanced, Transactional, and

Connected in 2010. Since then, the UN has shifted the focus of its objective to a more

holistic approach to capture a country’s performance in a single internationally-

comparable value.

43

2.4.2 EA in Developed Countries

The purpose of EA is to have a holistic approach in the development of an

integrated design (Saha, 2009). Having this in place will lead to better government

services (Peristeras & Tarabanis, 2000). To make the most of ICT, the United States

government received a mandate from Congress to reform the management of

Information Technology in their government. The mandate was called the Clinger-

Cohen act (The US Congress, 1996). In order to fulfil this law, the US Chief

Information Officer (CIO) introduced the use of EA into its e-Government. They

developed the Federal Enterprise Architecture Framework (FEAF) as a common

framework to be used in the US government (CIO Council, 1999). Since then, other

developed countries have adopted EA in their e-Government strategy.

2.4.2.1 The United Kingdom

The United Kingdom (UK) government acknowledged the importance of

technology in transforming its government. Hence, in 2005, “the transformational

government strategy enabled by technology” was introduced (Cabinet Office, 2005).

In this strategy, the UK government defined citizen centric ICT services that should

promote shared services in government. This in turn will enhance proficiency and

adaptability in the government to accommodate ICT enabled change. In order to

achieve this, the UK government established the eGovernment Unit (eGU)

(Liimatainen et al., 2007). From this unit, the UK government published the cross-

Government Enterprise Architecture (xGEA) in 2005 (Cabinet Office, 2010; Chief

Information Officer, 2005; Liimatainen et al., 2007). Since then, the UK government

refers to this xGEA in gaining sustainable alignment of business and ICT functions

(Cabinet Office, 2010).

44

2.4.2.2 Singapore

Singapore is a small country and considered to be the most advanced country in

the South East Asia region. Although Singapore has limitations in their resources, they

have successfully utilized technology to enhance their economy (Chua, 2012; ICT

Working Group, 2002). The Singaporean government is politically committed to

enabling ICT in their public services (Chua, 2012). Evidence of this commitment is

that in 2006 the Singapore government started to adopt EA (Pheng & Boon, 2007).

The development of the Singapore Government Enterprise Architecture (SGEA)

(Hanafiah & Goodwin, 2011; Pheng & Boon, 2007) was based on the US FEAF.

SGEA was completed with the inclusion of the Methodology for Agency ENTerprise

Architecture (MAGENTA) in 2007 (Pallab Saha, 2009).

2.4.2.3 The Republic of Korea

The Republic of Korea paid significant attention to e-Government development.

In order to enhance e-Government, the government started to develop Government

Wide Enterprise Architecture as their EA program in 2003 (National Information

Society Agency, 2011). By adopting EA, the Republic of Korea government expected

that it would overcome inter-department and inter-ministry integration issues (Joon,

2009). Although the EA program was started in 2003, the legislation related to EA was

not available until 2005 (National Information Society Agency, 2011). Later, this EA

law was formally merged into the e-Government act in late 2009.

2.4.2.4 Australia

In order to enhance e-Government initiatives, in 2007 the Australian government

decided to adopt the US FEAF in its EA program (Hanafiah & Goodwin, 2011). This

EA was called the Australian Government Architecture (AGA) (AGIMO, 2007). The

development of AGA was completed with the provision of the Business Reference

45

Model (BRM) details in 2009 (AGIMO, 2009; Hanafiah & Goodwin, 2011). With this

in hand, the government expects that EA will be able to deliver significant

improvement in e-Government systems.

2.4.3 Significance of EA to e-Government

This study measures the significance of EA in the above mentioned countries by

comparing their EA activities to their e-Government by analysing the UN data in the

e-Government index. It can be inferred that the overall e-Government index depicts

the snapshot of the e-Government development over time. As can be seen in Figure

2.6, in the first index published, the Republic of Korea e-Government was measured

slightly under the Singapore e-Government. However, the Republic of Korea

successfully gained the first rank with 0.879 out of 1.000 in the overall index in 2010.

They retained their position by achieving 0.928 in 2012.

Figure 2.6. Overall e-Government Index among observed countries

Although the other observed countries have not reached the same level as the

Republic of Korea, they either remained steady or gradually improved their ranking

2003 2004 2005 2008 2010 2012

Australia 0.840 0.838 0.868 0.811 0.786 0.839

Republic of Korea 0.737 0.858 0.873 0.832 0.879 0.928

Singapore 0.746 0.834 0.850 0.701 0.748 0.847

The United Kingdom 0.814 0.885 0.878 0.787 0.815 0.896

0.600

0.700

0.800

0.900

1.000

Overall e-Government Index

46

over time. All observed countries encountered a similar decline in 2008, the year when

the UN significantly modified its model (The United Nations, 2008). The Australian

e-Government index was considered steady from 2003 to 2008. In 2010, they

experienced a decline but successfully rebounded to 0.839 in 2012, two years after the

completion of their EA (AGIMO, 2009). Singapore’s e-Government and the UK’s e-

Government showed similar results. They remained steady from 2003 to 2005, but

declined in 2008. Notwithstanding, they managed to gain better results in 2010 and

2012.

The web measure/online service index, one of the three variables used to

measure e-Government, shows similar results to the overall e-Government index as

can be seen in Figure 2.7. The Republic of Korea achieved the maximum mark, 1.000

for this variable in 2010 and again in 2012 when it was joined by Singapore with a

perfect score. Australia and the UK have not yet achieved the same results but are

trending in the same direction by gradually increasing since 2008.

Figure 2.7. Web Measure/Online Service Index

2003 2004 2005 2008 2010 2012

Australia 0.812 0.830 0.904 0.753 0.765 0.863

Republic of Korea 0.607 0.946 0.977 0.823 1.000 1.000

Singapore 0.703 0.969 0.996 0.612 0.686 1.000

The United Kingdom 0.777 0.973 0.996 0.692 0.775 0.974

0.600

0.700

0.800

0.900

1.000

Web Measure/ Online Service Index

47

The achievements in the e-Government development indexes are in line with the

achievements in EA development. The Republic of Korea started their EA

development in 2003. They successfully enacted EA into law in 2005 and merged their

EA law into the e-Government act in 2009. Australia accomplished their EA in 2009

and started to gain steady growth in their e-Government index in 2010. Although

Singapore and the UK experienced a decline in the indexes after they enabled EA in

their e-Government in 2007 and 2005, they have made gradual incremental growth for

their overall e-Government index since 2008.

In addition, the Singapore e-Government experience shows that they have

positively minimized problems related to government interoperability. Saha (2009)

argues that since adopting EA, the Singaporean government has been able to develop

whole e-Government solutions that are comprised of modular services. Similarly, the

Republic of Korea positively reduced fragmented government services in their e-

Government system since the adoption of EA (National Information Society Agency,

2011).

2.5 EA in Developing countries

This study could not find any electronic documents as evidence for the presence

of EA in governments of developing countries. Although, some governments have

started to promote open government (McDermott, 2010), there remains many

government documents inaccessible due to secrecy and privacy concerns. Where

documents are available, such as presidential decree number 3/2003 for the Indonesian

e-Government implementation strategy, they do not necessarily reflect EA. For

example, the presidential decree makes use of architecture framework terminology,

but this is not viewed as an EA artefact by this study. This study could not map the

48

architecture framework for e-Government development in Appendix 4 of the decree

to any available EA framework such Zachman or TOGAF. Therefore, further study

needs to be done in relation to EA in the e-Government implementation in Indonesia.

Such studies also need to consider characteristics of Asian countries’ culture.

2.6 Asian Countries’ Culture Characteristics

Myers & Tan (2003) argue that culture is essential in the information systems.

However, globalisation has led a society to live inclusively (Walsham, 2002).

Walsham (2002) further argues that other cultures can influence other society.

Therefore, national culture is considered not to completely align with the territorial

boundaries. However, Tuunanen and Kuo (2015) found that there are differences

between participants from different cultural settings or locations regarding their

information system requirements. Therefore, understanding culture characteristics is

important.

A research from Trompenaars and Woolliams (2011) shows characteristics of

Asian culture. They found that developed and developing countries have different

characteristics. Indonesia, like other Asian and developing countries, is considered to

be a relationship-centred country rather than a rule-centred one (Trompenaars &

Woolliams, 2011). In a relationship-centred country, relationships are respected more

than abstract rules.

In their study, Trompenaars and Woolliams (2011) specifically mentioned

Indonesia as a communitarian country. In communitarian countries, errors were

considered as a group rather than an individual. Other characteristics of developing

countries can be seen in Table 2.4. The cultural characteristics of Indonesia, as a focus

49

country for this study, motivated the applied mixed methods approach presented in

Chapter 4 to achieve sound findings.

Table 2.4. Culture Characteristics (Trompenaars & Woolliams, 2011)

Characteristics Developed

Countries

Developing

countries

What’s more important Rules or Relationships? Rules Relationships

Who’s responsible to the failure? Individual Team

Do We Grant Status According to

Performance or Position?

Performance Position


This chapter found very few studies using EA as a theme in e-Government

research. Furthermore, this study found there is a relationship between EA adoption

and the performance of e-Government systems. Some governments in developed

countries such as the United States, the Republic of Korea, the United Kingdom,

Singapore and Australia appeared to have gained benefit from EA adoption in their e-

Government strategies. Replicating and extending the research by Shekkerman (2004)

by using more data series, it can be perceived that there is a strong correlation between

EA activities and e-Government system development. However, this study could not

find any evidence showing governments and government agencies of developing

countries, like the Indonesian Treasury, having adopted EA for their e-Government

systems. In order to have a further investigation on the Indonesian Treasury e-

Government systems, this study proposed a research model as seen in Chapter 3.

50

RESEARCH MODEL

3.1 Overview

The objective of this chapter is to develop a research model to be used in this

study. The literature review of Enterprise Architecture (EA) and e-Government in

Chapter 2 found that there is a correlation between the two. The relationship between

EA and e-Government is further investigated in Section 3.2. The benefits of EA are

described in Section 3.3. It is followed, in Section 3.4, by presenting how e-

Government performance is measured. Section 4.5 discusses an alternative EA

maturity model. Section 3.6 refines the EA Benefit Model (EABM) (Tamm et al.,

2011) by using EA maturity and SERVQUAL to be used to depict the correlation

between EA and e-Government performance. Hypotheses are developed based on the

proposed model in Section 3.7.

3.2 The relationship between EA and e-Government

It has become common to use EA for e-Government in order to overcome

government transformation issues that often neglect the importance of interoperability

(Wu, 2007). Waseda University of Japan have been including EA as one of the

parameters in its recent international e-Government ranking (Obi, 2012). From this, it

can be inferred that government are becoming increasingly aware about the importance

of interoperability of e-Government systems. Ignoring EA could restrict collaboration

between government institutions (Guijarro, 2007; Hjort-Madsen, 2006).

EA is believed to be well suited to enabling interoperability (Chen & Daclin,

2006, 2007; Daclin et al., 2006; Hjort-Madsen, 2006). Moreover, EA may also be used

to establish alignment between business and technology (Gregor et al., 2007; Martin,

51

2005; Pereira & Sousa, 2005; Seppanen et al., 2009; Shah & Kourdi, 2007). EA

focuses on both the technological and non-technological points of view (Finkelstein,

2006; Zachman, 1997; Zachman & Sowa, 1992).

3.3 Enterprise Architecture Benefits

EA is used “for the management of enterprise change. If enterprise architecture

is not used, there are only three options for managing enterprise change: by trial and

error; by reverse engineering; or by going out of businesses” (Zachman, 2003 cited

in Finkelstein, 2006, p. 9). Consequently, the use of EA is recommended as it can be

used not only to express the objectives of an enterprise but also to manage the change

required to achieve those objectives.

Several researchers have identified benefits of EA such as management of

change (Finkelstein, 2006; Zachman, 2003), alignment of business and ICT (Aversano

et al., 2012; Gregor et al., 2007; Pereira & Sousa, 2005; Wegmann et al., 2005), and

cost effectiveness (Finkelstein, 2006; Lankhorst, 2009; Ross et al., 2006; Zachman,

2003). Unfortunately, such claims are lacking in empirical evidence (Tamm et al.,

2011).

This study identified several models used to analyse EA benefits. First, van

Steenbergen et al. (2011) developed and validated a model to determine practices that

lead to EA benefits. Second, Lange, Mendling, and Recker (2012) developed and

validated the EA benefit realisation model (EABRM). Third, having conducted a

systematic review across an extensive array of articles about the benefits of EA, Tamm

et al. (2011) proposed the EA Benefits Model (EABM).

The first and second models defined EA to be a dependent variable. Van

Steenbergen et al. (2011) found that there are four practices that lead to good EA:

52

economic sector, project conformance to EA, choices in EA being explicitly linked to

business goals, and organised knowledge exchange between architects.

Similarly, in the EABRM, Lange et al. (2012) define factors influencing EA

benefits. The final model defined three independent variables and three mediating

variables. The independent variables are EA product quality, EA infrastructure quality,

and EA service delivery quality. The mediating variables are intention to use,

satisfaction, and EA culture. Lastly, the model defined EA net benefits as the

dependent variable.

The last model, the EABM, defines EA quality as the independent variable that

informs organisational benefit. In the model, Tamm et al. (2011) suggest that there is

no direct benefit resulting from EA to the organization. Instead, EA generates

intermediate factors that potentially lead to organizational benefits.

These intermediate benefit enablers exist between EA quality and the

organization benefits. The dotted lines in Figure 3.1 point from EA quality to resource

portfolio optimization and resource complementarity. These variables are dependent

on the partial creation of EA. Tamm et al. (2011) further argue that information

availability is partly dependent (indicated by the thin, solid line) on the existence of

EA. Finally, organizational alignment is achieved as a result of EA development

activities. This heavy dependence on EA activity is depicted by the thick, solid line.

53

Figure 3.1. The EABM (Tamm et al., 2011)

Tamm et al. (2011, p. 150) define each variable in the EABM as follows:

1. EA Quality is the degree to which the EA provides a clear vision of the future

system that is well aligned with organization goals. It is the only independent

variable in the model. There are two ways of assessing the quality of EA:

a. Directly evaluate products such as EA documentation.

b. Uses the EA development process as a proxy.

2. Organizational Benefits is defined as the outcomes that contribute directly to

organizational performance.

3. Organizational alignment is the extent to which an organization’s subunits

share a common understanding of its strategic goals, and contribute towards

achieving these goals.

Benefit Enablers

Enterprise Architecture Quality

Organisational Alignment

InformationAvailability

Resource PortfolioOptimisation

ResourceComplementarity

Organisational Benefits

Large Extent

Partly contingent

Dependent on at least partial of EA plans

Strength of the relationship

54

4. Information Availability is the extent of useful, high-quality information

accessible to organizational decision makers.

5. Resource portfolio optimization is the extent to which an organization

leverages its existing resources, invests in resources that target performance

gaps, and minimizes unnecessary investments in duplicated resources.

6. Resource complementarity is the extent to which the organization’s resources

synergistically support the pursuit of its strategic goals.

In light of the above, the EABM could be used to illustrate the relationship

between EA and e-Government performance. The EABM depicts the effect of EA

which is considered to be abstract as it is high level documentation as defined by

several researchers starting from page 34, on organisational benefit. Furthermore, it

shows that the result of having EA does not directly give benefit to the organisation.

The effect is rather mediated by four factors.

3.4 Measuring e-Government Performance using SERVQUAL

The performance of e-Government is in line with its maturity (Peters et al.,

2004). Academia, private consultants and non-profit organizations have proposed

ways to measure maturity. Peters et al. (2004) identified three groups of instruments

to measure e-Government: stage models, service quality, and performance indicators.

In the stage model grouping, there are some models that are already available to use,

e.g., a four-stage model (Layne & Lee, 2001), and a new model proposed by Siau and

Long (2005) resulting from synthesizing several models.

This study regards SERVQUAL as being appropriate for capturing users’

perceptions of e-Government systems performance. Constructive reviews of the

55

literature, presented in Chapter 2, resulted in this selection. Gorla (2012) argues that

SERVQUAL can be used to measure expectations and actual performance of a system.

Furthermore, SERVQUAL has been validated to measure the quality of information

systems in several developed countries (Jiang et al., 2002; Kettinger & Lee, 1994;

Landrum et al., 2009; Myerscough, 2002). However, this study could not find any

similar study of using SERVQUAL in developing countries. A study in the field of

Information Systems from the United Arab Emirates University was based on data

collected from CIO’s in North America (Gorla, 2012).

The SERVQUAL comprises five dimensions or factors. Parasuraman et al.

(1988) defined the factors as follow:

1. Tangibles: Physical facilities, equipment, and appearance of personnel

2. Reliability: Ability to perform the promised service dependably and

accurately

3. Responsiveness: Willingness to help customers and provide prompt

service

4. Assurance: Knowledge and courtesy of employees and their ability to

inspire trust and confidence

5. Empathy: Caring, individualized attention the firm provides its

customers

Studies in the Information Systems context adopted the model differently.

Landrum et al. (2009) used all dimensions in their research and concluded that

“SERVQUAL the instrument is not homogeneous, and some dimensions potentially

matter more to users than others”. SERVQUAL was combined with user satisfaction

with the information services function (USISF) (Kettinger & Lee, 1994; Myerscough,

56

2002). Jiang et al. (2002) excluded the tangibles factor in their research. Hence, the

dynamic nature of SERVQUAL is noticeable.

This study validates the use of SERVQUAL in developing countries. The model

is used to measure users’ perceptions about the performance of the Indonesian

Treasury e-Government systems. This study was aware of the unstable dimensionality

of the SERVQUAL (Van Dyke et al., 1997). As such, the study followed Myerscough

(2002) by administering the survey to users within the Indonesian Treasury to manage

the instability of SERVQUAL dimensions. This study generated alternative

dimensions for its SERVQUAL by using Exploratory Factor Analysis (EFA).

3.5 Developing an Alternative EA Maturity Model

It is important to have an instrument to measure the effectiveness of EA. One

way of doing that is by using EA maturity. As stated in Chapter 2, current EA maturity

models have apparent strengths and weaknesses. Using the meta-synthesis approach

(Noblit & Hare, 1988), proposed an alternative maturity model.

3.5.1 The meta-synthesis approach

According to Schreiber et al. (1997) meta-synthesis is “the aggregating of a

group of studies for the purpose of discovering the essential elements and translating

the results into the end product that transforms the original results into a new

conceptualization”. The meta-synthesis approach works through an interpretive

translation and comparison of some studies in a certain topic.

Schreiber et al. (1997) refers to the Noblit and Hare (1988) meta-synthesis

approach. It was Noblit and Hare who coined the meta-synthesis approach in 1988. It

involves a seven-step approach. The steps are: getting started, selecting relevant

57

studies, reading the studies, reviewing the related literature, translating the studies into

one another, synthesizing translations, and presenting the findings.

3.5.2 Meta-synthesis research process

Following the seven-step Noblit and Hare meta-synthesis approach, the

following processes have been followed in this research:

1. Getting started. In this stage, the objective of this research is to analyse EA

maturity models.

2. Selecting relevant studies. This research used the Internet to search for relevant

studies to this topic. Having used relevant databases in the systematic search

of the Internet, this research identified six EA maturity models. These models

were developed by a private enterprise, a government body, a non-profit

organization, an individual researcher and a well-recognized university.

3. Reviewing the related literature. Six EA maturity models were reviewed as

seen in Table 3.2. Repeated analysis was done to ensure that the details of the

entire model were investigated. This is a fundamental process to ensure

sufficient literature exploration.

4. Determining how the studies are related. This step and the following one are

the essential steps in the meta-synthesis approach. Initially, this stage identifies

similarities and differences among the models. First, the selected models were

investigated for similarities. They all have maturity levels for EA related

initiatives. Most of the models started their maturity level from the non-

existence of EA. The Ross et al. (2006) maturity model does not use the same

approach. It starts with the relationship between ICT and business. The six

models differ in terms of viewpoints. Some models were developed and

58

influenced by a particular EA framework, for example the Enterprise

Architecture Management Maturity Framework (EAMMF) and the US Office

of Management and Budget (OMB)’s Enterprise Architecture Assessment

Framework. The level of information carried by each model is also different.

The Institute For Enterprise Architecture Development (IFEAD)’s maturity

model has excessive detail in the measurement criteria, yet shifting strategy

and diagnostic tools are missing. The reviews of each model are summarized

in Table 3.1.

Table 3.1. EA Maturity model comparison

Maturity model Level

completeness

Shifting

strategy

Measurement

Details

Usability Diagnostic

tools

Enterprise Architecture

Maturity Model (NASCIO, 2003)

Not

Given

Sufficient General

Gartner’s IT Score (Burke, 2010) Not

Given

Sufficient General

GAO’s EAMMF (Hite, 2003) Brief Sufficient Specific

Four stages architecture

maturity (Ross et al., 2006)

Available Fair General

The IFEAD’s E2AMM v2 (IFEAD,

2004)

Not

Given

Rich General

OMB’s EAAF (Office of

Management and Budget, 2005)

Not

Given

Sufficient Specific

5. Translating the studies into one another. The objective of this step is to make

a comparison between models. One way to do this is by associating major

concepts, similarities and differences. At this step, this research investigates

the relationships between the models.

59

The research discovered there are six levels of relationship among models. The

relationships are depicted in Table 3.2 and Table 3.3. The relationships are

represented by a combination of a letter and a number, for example Gartner’s

model is represented by the letter G and its maturity level is represented by the

number 1 to 5. This representation is also applied to the other models.

It can be perceived from Table 3.3 that each level of the analysed models shows

a strong relationship to the level(s) of the other models. For instance the first

two levels of US GAO’s EAMMF are similar to Gartner’s first level. The initial

level of the Gartner model is an absence of EA and the establishment of an EA

team. Consequently, the Gartner model combines two different indicators, i.e.,

absence and building awareness. Further, this analysis found that the Gartner’s

top maturity level was not represented in the US GAO’s EAMMF. This means

that the Gartner model identifies the augmentation value of EA whereas the US

GAO’s model does not. Some notable relationships among the models can also

be seen where EA has been accepted in the enterprise, such as IFEAD’s level

3 and Ross et al.’s initial level. This analysis also discovered that all reviewed

models include maturity levels at which EA application begins and where there

is continuous alignment of business and technology.

6. Synthesising translations. At this stage, the research synthesises the

translations by showing the distribution of maturity levels. The research

reviewed the diagnostic tools, shifting strategies, measurement criteria,

artefacts and activities of each model. Comparison of the reviewed models can

be seen on Table 3.1. The model proposed by this study, as a result of the

review of extant models, offers identification of maturity level and strategies

for shifting from one level to another.

60

Table 3.2. Comparison Among Maturity Models

Level Gartner’s (G) NASCIO (N) The US GAO (UG) Ross et.al (R) The IFEAD (I) The US OMB (O)

1 Name Initial Level 0-No Program Stage 1 Business Silos Level 0 (No EA extension)

Undefined

Definition No Formal EA yet No documented architecture

No EA initiatives Focus on local context No Extension of EA No EA is available

Characteristic EA team is created Reliant on individual skills

EA awareness is built ICT automates specific business processes

Insufficient awareness between parties

No indication

2 Name Developing Level1-Informal Program

Stage 2 Standardized Technology

Level 1 (Initial) Initial

Definition Struggling to be successful

Informal Architecture Defined

Recognise the need of EA

Providing ICT efficiencies: business drives technology

Unforeseen participation within parties

Incomplete practices of EA

Characteristic Governance is clear Still Reliant on individual skills

EA management foundation is formed

ICT automates local business processes

Awareness brings initial principle of alignment

Informal and ad-hoc EA processes

3 Name Defined Level 2-Repeatable Program

Stage 3 Optimized Core Level 2 (Under Development)

Managed

Definition EA is in place Confirmable base architecture

Developing the architecture

Shift from local view to enterprise view

Visible awareness for the needs of partnership

High level definition of EA is available

Characteristic Measurement metrics are available

The standards have been used to confirm the performance

Framework, methodology, tool and plans are established

ICT enables enterprise achievements through optimizing reusable data and business platforms

Involving more parties in the program

EA processes are planned and managed

61

4 Name Managed Level 3-Well Defined Program

Stage 4 Business Modularity Level 3 (Defined) Utilized

Definition EA is delivering value & repeatable

Well defined architecture

Completing the EA Continuously refining digitized processes

Parties involved adequately in collaboration and information exchange

Some accomplishments refer to EA

Characteristic Integrated Processes Performance indicators are regularly monitored

EA has been approved Core system is linked with other internal and external systems through interfaces

High level officials amongst parties aware of the benefit of the program.

EA initiatives are documented, tacit and utilized

5 Name Optimizing Level 4-Managed Program

Stage 5 Level 4 (Managed) Result-oriented

Definition EA is driving the change

Performance indicators are analysed and acted upon

Leveraging EA High level officials review the program periodically.

EA processes are assessable

Characteristic Good stakeholder understanding

Indicators are used to forecast future capabilities

EA is used to ensure interoperability

Governance arrangement and management are available

Measured against performance standards

6 Name Level 5-Continuously Improving Vital Program

Level 5 (Optimized) Optimized

Definition Fully fit business and technology to the organisation vision

High level officials intensively involved in the optimization process

Constantly enable business improvement

Characteristic Continuous improvements occur.

Measurement metrics are ready to manage affected environment

Accountable cost spending to support productivity, saving, and service quality

62

Table 3.3. Illustrating relationship between one model to another

Gartner (G) NASCIO (N) The US GAO (UG)

Ross et al. (R) The IFEAD (I) The US OMB (O)

Gartner (G) - - - - - -

NASCIO (N) N0, N1 G1

N2G2

N3G3

N4G4

N5G5

-

The US GAO (UG) UG1,UG2G1

UG3G2

UG4G3

UG5G4

NULLG5

UG1 N0

UG2N1

UG3N2

UG4N3

UG5N4, NULLN5

- - - -

Ross et al. (R) NULLG1,G2

R1,R2G3

R3G4

R4G5

NULLN0,N1

R1,R2N2

R3N3,N4

R4N5

NULLUG1-3

R1,R2UG4

R3UG5

R4NULL

- - -

The IFEAD (I) I0,I1G1

I2G2

I3G3

I4G4

I5G5

I0,I1N0

I2N1

I3N2

I4N3

I5N4

NULLN5

I0UG1

I1UG2

I2UG3

I3UG4

I4UG5

I5NULL

I0-2NULL

I3R1,R2

I4R3

I5R4

- -

The US OMB (O) O1,O2G1

O3G2

O4G3

O5,O6G4

NULLG5

O1 N0

O2N1

O3N2

O4N3

O5N4

O6N5

O1UG1

O2UG2

O3UG3

O4UG4

O5,O6UG5

O1-3NULL

O4R1,R2

O5,O6R3

NULLR4

O1I0

O2I1

O3I2

O4I3

O5,O6I4

NULLI5

-

63

7. Expressing the findings. In this step, this research documents its findings as

text and diagrams. It was discovered that each model has apparent strengths

and weaknesses. Ross et al. (2006) presented a concise model in which the

alignment of business and ICT was already in place. They presented relevant

examples to show how this model is used. Therefore, their model is relatively

easy to follow and links with real, well known industries. Since their model

mainly focuses on an industry that has already began EA initiatives, new EA

entrants will not find their model suitable for their purposes as it is not useful

for working out how to put EA in place. Despite rigid measurement elements,

E2AMMv2 does not provide any diagnostic tools that will help in using this

model. Gartner’s model is concise and relatively easy to follow. Since it was

developed only to measure the maturity level, insufficient attention is paid to

the other criteria used in this research. Both the US GAO and the OMB focus

on measuring US government EA initiatives. To date, evidence has not been

found showing that this model has been used in other industries.

In light of the above, having linked the maturity levels for the various models

and considering the strengths and weaknesses of each model, this study proposes an

alternative maturity model. This model is built upon simple yet comprehensive

principles. This model is summarized in Table 3.4. This model is called the six As EA

maturity model. The benefits of this model are:

1. It refines the existing models that are not fully comprehensive for covering the

criteria used in this research.

2. It works for various organisations from early stage implementation of EA to

those with a mature EA. Current models are not entirely useful for EA adopters.

3. It provides a perceptible strategy for moving from one level to another.

64

Table 3.4. The six As EA maturity model

Lvl Level Name Indicator

1 Absence EA does not exist

2 Awareness Awareness is built, high ranking managerial approval is gained

3 Acceptance EA organization is established, involvement of more internal parties

4 Application EA is used as reference in aligning Business and ICT

5 Alignment Continuous alignment of Business/ICT is regularly conducted

6 Augmentation EA value is extended to indirect stakeholders

It is likely that any enterprise interested in developing and deploying an EA will

be able to utilize this model. A brief explanation of each level in this model is as

follows:

Absence. At this level, the organization does not recognize the need for an EA

despite the level of ICT usage in the business. Typically, in managing change, the

organization would use the trial and error approach. To shift from this level, it is

important to have someone, either internal to the organization or an external

consultant, influence high-level officials by showing the importance of EA. This

achievement is important because EA will be more effective if applied in a top-down

fashion (Finkelstein, 2006).

Awareness. Following the acceptance of EA by top-level officials, awareness of

EA should then be extended to relevant parties. An initial ad-hoc committee can be

used to start formulating or researching the EA framework and methodology to be

used.

Acceptance. At this level a formal unit that is responsible for developing and

maintaining EA has been established. It will include the involvement of other

stakeholders who will be directly affected by the EA. A formal model of EA should

65

also be available at this stage. The alignment of business and ICT has not been

established however.

Application. One feasible indicator at this level is that Business and ICT

transformations refer to the EA. EA becomes the main resource in fulfilling the needs

of the core system. This core system is represented by digitizing core data or business

processes within the organization (Ross et al., 2006). As enabling EA is a continuous

process (Gregor et al., 2007), regular review of the architecture will be needed to

satisfy supporting systems.

Alignment. Internal, core systems that support the running of the business are

developed with the guidance of EA. Business organization will also then be adjusted.

In order to move to the next level, it is essential to create a channel by which external

stakeholders, who are indirectly affected by the systems, can easily input their

concerns or views. This will enhance the systems to fit with their needs.

Augmentation. At the top level of this model, internal stakeholders and indirect

parties have gained the benefit of EA enabled systems. They have sufficient access to

propose improvements to the systems.

3.6 Refining the EABM

Although there are many studies with regard to the use of EA in government

(Janssen & Kuk, 2006; Liimatainen, 2008; Liimatainen et al., 2007; Martin, 2005;

Saha, 2010a; Seppanen et al., 2009), very few studies have successfully demonstrated

the value of EA to government (Liimatainen, 2008). Tamm et al. (2011) argue that

although there is no direct benefit of EA to an organization, the EABM displays the

relationship between EA and organizational benefit.

66

Government, the world’s largest and most complex organization (Saha, 2010a),

could adopt the EABM to find out if EA delivers benefit to their organization. One

significant benefit that needs to be identified is whether or not the e-Government

initiative is performing as expected. E-Government performance can then be used as

a variable to measure the benefit. Therefore, this study argues that EABM could be

adopted to define the relationship between EA and e-Government performance.

This study proposed EA maturity as the measure by which the quality of EA be

determined. In the EABM, Tamm et al. (2011) mention that EA quality could be

assessed in two different ways: assessing the final products of EA and using the

process of EA development as a proxy. These assessment methods are also used in EA

maturity (Banger, 2008; Burke, 2010; IFEAD, 2004; NASCIO, 2003) to measure the

effectiveness of an EA program within government or private institutions. Therefore,

the outcome and the process of EA maturity assessment are deemed similar to EA

quality as proposed in the EABM.

Although two variables have been identified and modified to investigate the

relation of EA and e-Government, this study needs to map the benefit enablers in the

EABM and the dependent variables that lead to e-Government performance.

3.6.1 Mapping EABM’s benefits and e-Government performance’s variables

The EABM proposed four variables that lead to organizational benefits. Tamm

et al. (2011) identified that it is possible to modify the grouping of the enablers in the

model due to the lack of empirical validation. The organizational benefit that this study

would like to discover is the e-Government performance. Currently, as mentioned in

an earlier section, various instruments to measure the e-Government performance are

already available.

67

As argued on pp. 54 - 56, in measuring e-Government performance, this study

adopts SERVQUAL. This study uses a modified SERVQUAL (Parasuraman et al.,

1988) from the field of Information Systems (Gorla, 2012).

To find theoretical relationships, the variance approach is used. The variance

approach reflects models that have different entity properties (Burton-Jones et al.,

2011). In this manner, it is essential to map the EA benefits and e-Government

performance to find out whether or not the EABM’s benefits and e-Government

performance’s variables fit. The mapping of these models is depicted in Figure 3.2.

The EABM defined four dependent variables that may influence the benefit of

EA (Tamm et al., 2011), while Parasuraman et al. (1988) define five variables to

measure service quality.

Organizational alignment in the EABM is used to measure how organizations

share a common understanding, which in turn improves reliability, responsiveness,

assurance and empathy.

Information availability is used to identify the usefulness and accessibility of the

output from e-Government. Consequently, this requires reliable, fully responsive and

assured IS. The last two variables, resource portfolio optimization and resource

complementarity, are used to ensure there will be efficient infrastructure investment.

These variables are tangibles. Insufficient infrastructure performance may lead to loss

of organizational empathy.

68





Reliability

Tangibles

Responsiveness

Empathy

Assurance

Figure 3.2. Mapping EABM’s Benefit Enablers and SERVQUAL variables

3.6.2 Proposed Model

Having reviewed variables in the EABM and SERVQUAL, it was found that

both models are considered to be complementary and appropriate, to an extent, to the

purposes of this study. Therefore, this study proposes a refinement to the EABM to

depict the relationship between EA and e-Government performance. Modified

instruments from Gorla (2012) for variables in SERVQUAL (Parasuraman et al.,

1988) could then be used to measure e-Government performance.

The maturity of EA drives the performance of e-Government through the

performance drivers adopted from the EABM. Therefore, the more mature EA, the

better performance that e-Government can deliver. Consequently, EA can be used to

improve the quality or performance of e-Government. In order to measure this, the

69

model depicted in Figure 3.3 is used. This study suggests the strength of relationship

in the model to be equal to all mediators because this study could not find any study

validating the strength of relationship between EA and performance enablers.

The degree of EA maturity is assessed through completeness of EA

documentation and its development process (NASCIO, 2003). Although this study

proposes an alternative EA maturity model, measurement details have not yet been

developed. Adopting NASCIO (2003), the following categories are used to measure

the maturity level of an EA program: administration to measure governance and

responsibilities; planning to gather the road map and execution plan; templates and

process used to measure the framework; blueprint is gathered from collection of

standards used; communication is assembled from education and distribution of EA;

compliance is collected through adherence among EA elements; integration is

compiled through management process engagements to exchange information; and

involvement to determine user participation in the development and deployment

process.

This study used modified SERVQUAL as suggested by Gorla (2012) to measure

the range of facilities used, how up to date is the hardware and software, engagement

of personnel and the operating hours of the IS department for the tangibles variable.

The reliability factor was measured through fulfillment of promises, dependability,

timely service delivery, and service commitment. Prompt responses, willingness, and

staff availability were used for the responsiveness variable. Some elements are used to

measure assurance such as: instilling confidence, feeling safe in using the system,

interaction between the IS department and users, and IS employees’ knowledge.

Empathy includes individual attention to users, personal approach to users, IS

70

department has the user’s best interest in mind and they understand user’s needs. These

variables will be used to measure the e-Government performance.

Figure 3.3. Relationship of EA and e-Government Performance

The first performance enabler in the model, namely organisational alignment

(OA), was used to measure common understanding (Bernard, 2012; van Steenbergen

et al., 2011) of the Indonesian Treasury strategic goals. In regard to OA, the study also

captured the presence of good communication between the business and ICT technical

parties (Kappelman, 2010; Pereira & Sousa, 2005) in developing the e-Government

systems. The manifestation of Business-ICT alignment (Lankhorst, 2004; Ross et al.,

2006) was collected as part of measuring organisational alignment.

The second enabler in the model, information availability (IA), was used to

investigate the information accessibility (Spewak & Hill, 1993). The study also

measured the quality of information generated by the systems (Bernard, 2012;

Performance Enablers

EA Maturity





E-Government Performance

71

Finkelstein, 2006; Venkatesh et al., 2007). In addition, data management (Finkelstein,

2006; Ross et al., 2006; Spewak & Hill, 1993) was captured to measure the information

availability.

The third enabler, resource portforlio optimisation, measured the detailed

resources used in terms of numbers and commonality in the Indonesian Treasury. The

presence of EA could minimise duplication of resources (Pereira & Sousa, 2005;

Zachman, 1997). In turn, it could reduce the cost of unnecessary ICT investment (Perks

& Beveridge, 2002; Ross et al., 2006; Spewak & Hill, 1993).

The last enabler in the model, resource complementarity, was used to measure

integration among enterprise resources (Bernard, 2012; Spewak & Hill, 1993;

Zachman, 1997). The presence of EA can promote interoperabilty (Alwadain, 2014;

Chorafas, 2001; Ross et al., 2006) among resources to support systems integration.

This study goes on to develop the research instruments as presented in Chapter

6 and Chapter 7 to measure the above mentioned dimensions. Such instruments were

then presented as a questionnaire as seen in Appendix B and a list of interview

questions as seen in Appendix C. All measurements were then used to validate the

hypotheses of this study.

3.7 Hypotheses

This study aims to investigate the effect of EA on e-Government performance.

The research model in Figure 3.3 shows that e-Government performance is not directly

affected by EA. Since EA is cosindered to be a high level design artefact (Bernard,

2012; Lankhorst, 2009), its effect is not seen instantly. As such, the reseach model

identified that the relationship between the EA and e-Government performance is

72

mediated by four dimensions, namely, organisational alignment, information

availability, resource portfolio optimisation and resource complementarity.

Hence, this study developed its hypotheses by using mediation analysis (Hayes,

2013; Iacobucci, 2008). Mediation analysis can better explain modelling a process and

be used beyond analysing data (Hayes, 2013; Iacobucci, 2008; Kenny et al., 1998).

Kenny et al. (1998) further contend that mediation analyses using structural

equation modelling for testing hypotheses to be better than ANOVA3. Moreover,

evaluating direct effects of EA on each of the mediators and direct effects of mediators

on e-Government performance could divert the aims of this study. To understand the

effect of each mediator in the model, each of the mediators was analysed separately

(Kenny, 2014; Kenny et al., 1998) using simple mediation analysis (Hayes, 2013;

Iacobucci, 2008).

The following hypotheses were then used to empirically validate the proposed

research model and its data from the Indonesian Treasury:

H1: Organisational Alignment mediates the positive effect of EA on e-Government

performance in the Indonesian Treasury.

Figure 3.4. Hyphothesis 1 of the study

3 Analysis of variance (ANOVA) is used to compare group means in statistical method (Pallant,

2011)

EA



73

H2: Information availability mediates the positive effect of EA on e-Government

performance in the Indonesian Treasury.


H3: Resource portfolio optimisation mediates the positive effect of EA on e-

Government performance in the Indonesian Treasury.


H4: Resource complementarity mediates the positive effect of EA on e-Government

performance in the Indonesian Treasury


3.8 Definition of Constructs

The framework used in this study consists of six constructs with the maturity of

EA being the independent factor. The other constructs are dependent variables:

Organisational Alignment, Information Availability, Resource Portfolio Optimisation,

Resource Complementarity and e-Government Performance. The definition for each

construct is summarised in Table 3.5.

EA

Information Availability


EA

Resource Portfolio Optimisation


EA

Resource Complementarity


74

Table 3.5. Definition of Constructs

Construct Definition Reference(s)

EA Maturity The quality of EA measured through

the maturity of EA and the activities

related to EA development.

(Hanafiah & Goodwin,

2012; NASCIO, 2003;

Tamm et al., 2011;

Venkatesh et al., 2007)

Organisational

Alignment

The extent to which organisations in

the Indonesian Treasury have

common understanding of its

strategic objectives, and contribute

toward these goals.

(Bernard, 2012;

Kappelman, 2010;

Lankhorst, 2009;

Pereira & Sousa, 2005;

Ross et al., 2006;

Tamm et al., 2011; van

Steenbergen et al.,

2011)

Information

Availability

The extent to which Indonesian

Treasury high-level officials

generate useful, high-quality

information.

(Bernard, 2012;

Finkelstein, 2006; Ross

et al., 2006; Spewak &

Hill, 1993; Tamm et

al., 2011; Venkatesh et

al., 2007)

Resource Portfolio

Optimisation

The extent to which the Indonesian

Treasury makes the most of its

existing resources, invests in

relevant resources to close

performance gaps, and minimises

duplicated resources.

(Pereira & Sousa,

2005; Perks &

Beveridge, 2002; Ross

et al., 2006; Tamm et

al., 2011; Zachman,

1997)

Resource

Complementarity

The extent to which the Indonesian

Treasury’s resources synergistically

support its strategic goals

(Alwadain, 2014;

Bernard, 2012;

Chorafas, 2001; Ross

et al., 2006; Spewak &

Hill, 1993; Tamm et

al., 2011; Zachman,

1997)

e-Government

Performance

Using SERVQUAL to measure the

extent to which the Indonesian

Treasury e-Government systems

performance meet the expected

goals.

(Gorla, 2012; Hanafiah

& Goodwin, 2014;

Parasuraman et al.,

1988)

75


This study developed its model by extending the EABM to depict the

relationship between EA and e-Government system performance. The SERVQUAL is

considered to be a complementary fit to measure the performance. The SERVQUAL

has already been empirically tested in various studies in the field of Information

Systems (Gorla, 2012; Kettinger & Lee, 1994; Myerscough, 2002; Van Dyke et al.,

1997). This study investigates the use of its model in a government institution. To

ensure the fitness of this model in the setting of a developing country’s government,

selection of appropriate research methodology is needed as presented in the following

chapter.

76

RESEARCH METHODOLOGY

4.1 Overview

This chapter covers the selection of a research methodology intended to validate

the research model. It begins with case study research in Section 4.2. Section 5.3

presents the research framework for this study. Data sources are identified in Section

4.4. This study uses a mixed methods approach as presented in Section 4.5 by

employing research techniques identified in Section 4.7. This chapter also discusses

the selection of respondents in Section 5.6. Data collection, coding and ethical issues

are presented in Section 4.8 and Section 4.9. This study’s research process is illustrated

in Figure 4.1 and its literature coverage in developing its research model in Figure 4.2.

4.2 Case Study Research

A characteristic of Information Systems (IS) research is the need to consider

rapid technological changes (Benbasat et al., 1987). As a dynamic subject, research in

the IS field requires a relevant research methodology. Case studies are one of the

methodologies that can cope with rapid technological change. The strength of case

study research is in its capability to reveal causal paths through richness of details

(Garson, 2013a).

The use of case study in the information systems field is common (see Cavaye,

1996; Gable, 1994; Gupta & Jana, 2003; Richard & Michael, 2002). Benbasat et al.

(1987) argue that:

Case study examines a phenomenon in its natural setting, employing multiple

methods of data collection to gather information from one or a few entities

(people, groups, or organizations).

77

Preparing the Study

Producing the research proposal; Reviewing relevant literature; Developing research model; Selecting research subject;

Obtaining ethics approval; Obtaining permissions from research site.

Collecting Data

Simultaneous Mixed-Methods

Quantitative

Using 6-point Likert scale paper-based questionnaire;Adopting simple stratified sampling, sub population

and quota sampling; Receiving 561 responses of

621 invited.

Qualitative

Interview

Using purposive sampling;

Interviewing 15 of 20 respondents.

Documentary

14 ICT related regulations;

Mailing systems

Observations

Treasury Service Office (TSO) in

Jakarta;

TSO in Tasikmalaya.

Analysing Collected Data

Quantitative

Collected Data were analysed using:IBM SPSS version 22 software and

IBM AMOS version 22

Qualitative

Collected data were analysed using NVivo version 10 software.

Presenting Findings and Thesis Writing

Discussing the findings based on the literature and research model; Ongoing literature review throughout;Presenting the significance; Writing up the Thesis.

Figure 4.1. The Research Process Of This Study

There are several definitions of case study. This study referred to the definition of case

study from Yin (2009). He defines case study as:

To understand a contemporary phenomenon in depth and within its real-life

context, especially when the boundaries between phenomenon and context are

not clearly evident.

The characteristic of case study:

78

Copes with the technically distinctive situation in which there will be many more

variables of interest than data points; relies on multiple sources of evidence,

with data needing to converge in a triangulating fashion (Yin, 2009)

Conducting case study research is at its best when: it is conducted in a natural

setting or on current events; is based on newly developed theory to understand the

research background; there is less control over the research events (Benbasat et al.,

1987).

This study is well suited to case study research since: Enterprise Architecture

(EA) and e-Government is a contemporary topic in major developing countries; it is

conducted upon a relatively new developed theory of improving e-Government

performance through EA based on the EA benefit model; and, there is limited control

over the research setting—the Indonesian Treasury.

To make the most of case study research, Garson (2013a) argues that the mixed

method approach is required to increase the quality of findings when generated from

a single case study. Therefore, this study employed a mixed method approach in

collecting the data to gain exceptional outcomes based on the research framework

described earlier in Chapter 4.

4.3 Research framework

In formulating a research model of the effect of EA on e-Government

performance, this study took a deductive approach. The deductive approach validates

theory as a result of a literature review (Creswell, 2009; Van Dinther, 2008). The

process of the deductive approach includes developing hypotheses, data collection,

and accepting or rejecting hypotheses.

This study included three terms in its theoretical research: e-Government,

Enterprise Architecture and the Indonesian Treasury, as reflected in Figure 4.2. Each

79

term is discussed in a separate chapter. Chapter 2 discussed the wide range of theories

in e-Government and Enterprise Architecture that led to answering the first question

in the study: internationally, has EA implementation demonstrated benefits to e-


Figure 4.2. Literature Coverage in This Study

Chapter 5 discusses several issues present in the Indonesian Treasury

particularly in its e-Government systems development. The aim of Chapter 5 was to

answer the other question on the list: What is the current state of the Indonesian

Treasury e-Government systems? How can it be improved? This was an important

chapter to understand the determinants in the research model.

This study relies on the determinants of the effect of EA on e-Government

performance derived from the EABM as discussed in Chapter 3. This chapter

Investigating ways to improve e-Government performance through

Enterprise Architecture.


The Indonesian Treasury

E-Government in Developing

Countries

Literature

Literature

Literature

80

responded to the question: how can the quality or the process of EA development affect

e-Government performance? At the end of these three chapters, the study proposes the

research model as seen in Figure 3.3.

The research model was empirically validated in the Indonesian Treasury to

answer the last question: How can EA be used to improve the quality of the Indonesian

Treasury e-Government systems? Hence, this study can answer its focal question: to

what extent can EA be used to improve the Indonesian Treasury e-Government

performance? The data-collection process in this study is described in the subsequent

sections.

4.4 Identification of the Data Sources

The Indonesian Treasury is comprised of three groups of offices: headquarters,

Treasury Regional Offices (TRO) and Treasury Service Offices (TSO). With the

different levels of hierarchy in the Indonesian Treasury, this study needed to find the

most appropriate sources of data in order to minimise bias. Leaders in the headquarters,

including the Director General, were identified to significantly contribute to this study.

TROs, wherein their former leaders’ roles had significantly contributed to the e-

Government systems development for the Treasury, contributed sound findings.

The TSOs are the main users of the Indonesian Treasury e-Government systems.

Thus, their views of current e-Government systems were needed to gather reliable data.

As well, other leaders’ within the Ministry of Finance, who were considerably

involved in the e-Government systems development for the Indonesian Treasury, were

also used as resources. In addition to people as the source of data, regulations and

electronic records were also used to contribute to the findings.

81

Hence, this study adopted both qualitative and quantitative exploration to meet

the characteristics of the particular data source. A qualitative approach, such as via

interview, is better than distributing questionnaires for high level officials at

headquarters and the TROs, due to their time constraints. The other levels of

respondents had more flexibility in managing their time. Both groups indicated

significant interest in contributing to the study through their responses.

4.5 Selection of Methods: Mixing Qualitative and Quantitative Methods

This study used mixed methods, a combination of qualitative and quantitative,

to answer research questions. One of the reasons for selecting mixed methods was the

research setting. As mentioned earlier in Chapter 2, Indonesia has distinctive

characteristics. Consequently, relying exclusively on one method may generate an

incomplete picture of the setting. In addition, Venkatesh et al., (2013) argue that IS

researchers can benefit from mixed methods design strategies because of the rapid

development of technology in ICT. Hence, the use of mixed methods in IS research

can significantly add to theory and practice.

Venkatesh et al. (2013) defines seven reasons for conducting mixed methods

research in IS, as can be seen in Table 4.1. The purpose of using mixed methods in this

study was to capture the complete picture of how the Indonesian Treasury e-

Government systems are performing. As a result, this study is able to generate

comprehensive findings.

The use of mixed methods can also improve the credibility of a study. Mixed

methods is also known as “triangulation”, “hybrid”, and “integrated” method

(Creswell, 2009). In mixed methods, data collection can either be done sequentially or

82

simultaneously (Johnson & Onwuegbuzie, 2004; Small, 2011; Venkatesh et al., 2013).

This study collected its data simultaneously for about six months from September 2013

to March 2014.

Table 4.1. Purpose of Mixed Methods (Venkatesh et al., 2013, p. 26)

No Purpose Description This Study

1 Complementarity Mixed methods are used in order to gain

complementary views about the same

phenomena or relationships.

○

2 Completeness Mixed methods designs are used to make

sure a complete picture of a phenomenon

is obtained.

√

3 Developmental Questions for one strand emerge from the

inferences of a previous one (sequential

mixed methods), or one strand provides

hypotheses to be tested in the next one.

○

4 Expansion Mixed methods are used in order to

explain or expand upon the understanding

obtained in a previous strand of a study.

○

5 Corroboration/

Confirmation

Mixed methods are used in order to assess

the credibility of inferences obtained from

one approach (strand).

○

6 Compensation Mixed methods enable compensating for

the weaknesses of one approach by using

the other.

○

7 Diversity Mixed methods are used with the hope of

obtaining divergent views of the same

phenomenon.

○

Multiple sources of data were used to ensure complete findings. Data from high-

level officials of the Indonesian Treasury were collected through semi-structured one-

on-one interviews. ICT related regulations both in hardcopy and electronic copy were

used as part of a document analysis. Since the Indonesian Treasury administers its

mailing systems electronically, this study also used its data to analyse the

communication between headquarters and other offices. This study used

83

questionnaires to capture responses from e-Government systems’ users in the TROs

and TSOs.

4.6 Selection of Respondents

This study applied a mixture of stratified random sampling and purposive

sampling (Fowler, 2014; Fuller, 2009) as seen in Table 4.2. Interviewees were selected

through purposive sampling (Clark & Creswell, 2011; Creswell, 2009). The

interviewees’ background, roles, and contribution to the e-Government systems in the

Indonesian Treasury were used to satisfy the purposive sampling. The information

about participants were gathered through publicly available documents on the

Indonesian Treasury website and based on the researcher’s experience. Each

participant’s profile and recent contributions to the e-Government systems

development were reviewed to achieve the objectives of this study.

Table 4.2. Respondent Selection Technique

Level TRO TSO Total invitation

per Office

Echelon 2 Purposive Sampling N/A 14

Echelon 3,

4 and Staff

Stratified random

sampling

Stratified random

sampling

1

The study used a stratified random sampling technique to invite participants to

complete a questionnaire. The size of the population is known. Thus, the selection of

the sampling method is manageable (Fuller, 2009). The use of these techniques should

fit the findings from the Indonesian Treasury. During data collection, the Indonesian

Treasury had 30 TROs and 177 TSOs as its operational offices. All offices were

invited. Every staff level in each office received one invitation. Although initially

4 Not all heads of TROs were invited. Only representatives of three regions were invited.

84

invitations to three different levels of officials were deemed adequate, this study let

selected respondents invite others to voluntarily contribute to the survey.

Interviews were mostly conducted in the headquarters. Three out of fifteen

interviewees were from the TROs. The three TRO participants demonstrated

significant contributions to the e-Government systems development. For instance,

Respondent 6 was involved in deploying the Intranet and the modern TSO and

Respondent 7 was involved in deploying the Treasury Single Account, which is

dependent on proper ICT systems. In addition, they represented three different regions

in Indonesia, i.e. Sumatera and Borneo, Java, and Bali and Eastern Indonesia.

In total, this study invited eight participants from Echelon 2 to be interviewed:

three from TROs and five from headquarters. The same number of invitations was sent

to Echelon 3 in the headquarters. However, they comprised two categories. Four

interviewees were from general roles in the headquarters and four interviewees had

ICT related roles. As for Echelon 1, this study invited four officials to take part in the

study.

4.7 Research Techniques

As mentioned in the previous section, this study uses multiple research

techniques as part of its preferred mixed method. Most of the research took place in

the Indonesian Treasury headquarters. Two out of three interviews with heads of TROs

were conducted in their offices. Another interview with the head of a TRO was

conducted in the headquarters when the interviewee was in Jakarta. Questionnaire

responses from the TROs and TSOs were collected through postage.

85

Table 4.3. Research Techniques and Sources of Data

No. Institution Technique Number of people involved

1. Headquarters Interview

Document Analysis:

Regulations

Mailing Systems

12 officials

3 staff

1 systems administrator

2. Treasury Regional

Offices

Interview

Questionnaire

3 officials

66 respondents

3. Treasury Service

Offices

Questionnaire

Observations

490 respondents

2 offices

The mixture of interviews, surveys and document analysis was conducted

simultaneously. Data collection started in September 2013 and was finished as planned

in March 2014. This study observed the e-Government system performance during its

peak period, the end of Indonesian government fiscal year in December.

4.7.1 Interview

The nature of the research subjects of this study, EA and e-Government, is that

they are generally driven by officials at different levels. Thus, their thoughts regarding

the research subject in the Indonesian Treasury were required. One way to do this is

through interview.

Despite some drawbacks such as bias, time, lack of anonymity, and cost, the

interviews were deemed relevant for this study. The shortcomings are manageable. All

interviews were transcribed. The results were sent to the interviewees to get their

feedback. Hence, the study was able to control the bias. A list of questions were sent

at least a month before the designated time for the interview. These questions were

used as guidelines in the meetings rendering the interviews as semi-structured. The

questions were generated based on identifiable data generation strategies proposed by

Schultze and Avital (2011).

86

Semi-structured interviews were chosen due to the following reasons: it was

unlikely to be able to get more than one opportunity to conduct an interview with the

high level officials in the Indonesian Treasury; and, the possibility for different

terminologies being used requiring time to synchronise understanding between

interviewer and interviewee. In addition, the semi-structured interview provides

flexibility in gathering the data from respondents (Myers & Newman, 2007; Newton,

2010; Schultze & Avital, 2011).

Interviews were all conducted in the interviewees’ rooms or a place nominated

by the interviewees. They were conducted individually and tape-recorded. Information

regarding the purpose of the study, a list of questions, consent, and anonymity were

provided in the invitation letter. The interviewer reiterated all those points to the

interviewees. Despite availability of consent forms, most interviewees preferred using

verbal consents rather than signing them. All interviews were conducted by one person

only to maintain consistency.

4.7.2 Questionnaire

In addition to the interviews, the study collected data from more respondents

within the Indonesian Treasury by using a questionnaire. The questionnaire comprised

six levels of categorical data based on a 6-point Likert scale (Croasmun, 2011; Lei

Chang, 1994; Likert, 1932) by excluding the neutral point (Guy & Norvell, 1977) to

minimise unexpected responses affected by culture (Lee et al., 2002). Questionnaires

provide data based on predefined variables and values used to analyse the research

subjects (Gray, 2004).

Benefits of using a questionnaire include assuring anonymity, being

inexpensive, being accessible and convenient (Achmad, 2012; Wheeldon & Ahlberg,

87

2011). Nevertheless, using a questionnaire has shortcomings such as being less

flexible, having a low response rate, and lack of control of environment. The study

managed these shortcomings through: carefully choosing respondents who might

foresee the benefits arising from the study; designing an attractive questionnaire; pre-

testing the questionnaire with statistical experts, PhD students from Indonesia and

elsewhere, and Indonesian Treasury employees who were undergoing post graduate

studies in Adelaide.

The study adopted paper-and-pencil questionnaires rather than computerised

questionnaires. Paper-and-pencil questionnaires are deemed appropriate based on a

recent study from Achmad (2012). He adopted both types of questionnaire in his study

at the Indonesian Treasury. He received a higher response rate from the paper-and-

pencil questionnaires than from the computerised ones. He sent invitations for

computerised questionnaires through e-mail and letter by providing the questionnaire

website address. He made the questionnaire available on the web and sent some of

them through e-mail attachments. Instead of repeating his approach of providing the

questionnaire via multiple mediums, this study selected paper-and-pencil

questionnaires only.

The study sent questionnaires to three different staff levels in the hierarchy of

the TROs and TSOs. This allowed responses to be captured from various backgrounds.

Questionnaires were sent through the postal system. To remain anonymous, each

respondent was given a postage paid, self-addressed envelope with which the

completed questionnaire could be returned.

88

The result of this careful preparation paid off. The study received a relatively

high response rate of around 90 percent; 561 responses were returned of 621

questionnaires sent as seen in Table 4.4.

Table 4.4. Survey Response Rate

Respondent’s Office Invited Responded5 Percentage

Treasury Regional Offices 90 66 73.33 %

Treasury Service Offices 531 490 92.28 %

Not declared 5

Total 621 561 90.33%

4.7.3 Documentary Evidence

The study collected relevant documents such as regulations, reports, systems

documentation, and administration mailing archive. All documents were held in the

headquarters. The documentary evidence was used to supplement the surveys and

interviews. Documents are considered to be inconspicuous, firm, rigorous, and far-

reaching (Gray, 2004). Hence, documentary evidence was deemed useful.

Nonetheless, documentary evidence has drawbacks, for instance, accessibility

(Achmad, 2012). The researcher’s background as a member of the staff of the

Indonesian Treasury assisted in being able to access required documents. The

regulations, current and previous e-Government systems’ documentation, reports

made by consultancy services and the Indonesian Treasury, financial data in hard and

electronic copies, computer specifications, and road maps were collected from the

Indonesian Treasury.

5 Based on respondent’s returning the questionnaire

89

4.7.4 Observations

E-Government systems are meant to improve the quality of services in the

Indonesian Treasury, specifically in the TSOs. The TSOs are the place where Treasury

related transactions occur. For example, the spending units send their spending orders

daily, designated banks and post offices pass their data for revenue transactions on a

daily basis as well. Thus, parties who are involved in e-Government systems

interactions can be seen in the TSOs. This study used observation to capture this.

Observation was deemed appropriate as it provides a chance to portray how the

research subject is behaving (Gray, 2004).

Two TSOs were selected: one in Jakarta and the other in Tasikmalaya,

approximately 250 kilometres south east of Jakarta. The TSO in Jakarta is considered

to be the busiest TSO in Indonesia with nearly half of the overall Indonesian budget

being handled in this office. The observation in this office was conducted for about

three weeks from the end of November to the middle of December 2014 a few weeks

before the Indonesian fiscal year officially ended at the end of December. The

Tasikmalaya TSO is a more moderate office. It was selected to support observations

made in the Jakarta office. Hence, the study spent only a week in the Tasikmalaya

TSO.

The observations were concentrated during the time spent processing spending

units’ requests. The processes, the queues, the timeliness of service, the interaction

between the TSO staff and spending units were monitored. Hence, the process of

spending authorisation, accounting data reconciliation, and revenue data handling was

observed.

90

4.8 Data collection, Coding and Analysis

Data were collected concurrently with surveys, collection of documentary

evidence, observations, and interviews all being conducted in the same period of time.

Individual interviews were audio recorded with notes taken to mark important

messages, gestures or expressions. Recorded audio and notes were then compiled into

text and stored in electronic forms. These were stored on the University’s network

drives as the ICT department regularly backs them up. A private external drive was

also used as extra backup storage.

Secondary data from documentary evidence were copied into electronic forms.

The electronic copies of qualitative data excluding data from observations were

transferred into NVivo. NVivo is software used to support qualitative data analysis

(Bazeley & Jackson, 2013; Lakeman, 2008). Interview transcripts were kept in the

original language (Indonesian). Following Gray’s (2004) suggestion for hastening and

improving the quality of data analysis, data were classified into specific codes

according to the type of documents, respondent’s rank, and purposes. The coding was

based on the research framework discussed in Section 3.6.2.

Following the research framework presented in Section 3.6.2, the 51 observed

variables in the quantitative data were reduced to six factors as seen in Table 4.5. Most

questions in the quantitative surveys used ordinal data. In addition to ordinal data,

questions to capture demography of respondents used scaled and nominal data. The

data were stored in SPSS6 and AMOS7 for further analysis.

6 Statistical Package for the Social Sciences (SPSS) is software package for statistical analysis

(Pallant, 2011). 7 Analysis of Moment Structures (AMOS) is a software that can be used to analyse statistical

modeling using graphical interface (Byrne, 2010)

91

The study applied factor analysis to assist with the aim of this study in finding

the correlation between EA and e-Government performance. Data collected into

observed variables were reduced to six factors. A combination of Exploratory Factor

Analysis (EFA) and Confirmatory Factor Analysis (CFA) was used to measure

correlations. Additional analysis was used using several measurements to ensure the

goodness-of-fit of the model.

Table 4.5. Dimension Reduction of Observed Variables

No. Dimension (Factor) Name Total Variables Used

1 e-Government Performance 5 sub factors (22 observed variables)

from the SERVQUAL

2 Organisation Alignment 5 observed variables

3 Information Availability 5 observed variables

4 Resource Portfolio Optimisation 5 observed variables

5 Resource Complementarity 5 observed variables

6 Enterprise Architecture 9 observed variables

The research framework model was converted into statistical modelling using

AMOS version 22. Further analysis was then based on Structural Equation Modelling

(SEM). SEM is used to test the hypothesised relationships between variables, both

observed and latent variables, through comparison of data and model (Byrne, 2010;

Hancock & Mueller, 2013; Hoyle, 2012).

This study analysed its qualitative and quantitative data separately as reported in

Chapter 5 and Chapter 7. The results are then mixed and discussed in Chapter 8.

Creswell (2009) argues that mixed method analysis can be run by:

Mixing of the two types of data might occur at several stages: the data collection,

the data analysis, interpretation, or at all three phases. (p.185)

92

Mixing means either that the qualitative and quantitative data are actually

merged on one end of the continuum, kept separate on the other end of the

continuum, or combined in some way between these two extremes (pp.207-208).

4.9 Ethical Issues

This study required interaction with humans as research participants. The

interactions with humans occur during interviews, collecting government documents,

and during observation. Indirect interaction with humans occurs during the survey

questionnaires. Therefore, prior to conducting data collection, this study sought ethics

approval from the Social and Behavioural Research Ethics Committee at Flinders

University. Ethics approval was received on 15 August 2013 with project number 6205

as seen in Appendix A.

During data collection, the researcher introduced himself as a research student

at Flinders University and as a staff member at the Indonesian Treasury and elucidated

the purposes of the study and the benefits of participating in the research. The

researcher then asked participants’ consent to record all interviews. The participants

were offered consent letters to sign, although, most interviewees chose to give verbal

consent.

The study also received permission from the Indonesian Treasury to collect

relevant documents on the research subject, to conduct observations in two offices and

to distribute paper questionnaires. The cover letter of each questionnaire informed the

respondents that the completion and returning of questionnaires was considered to be

their approval to participate in the research. Therefore, no explicit consent letter was

sent to the respondents of the survey questionnaire.

93


This chapter presented supporting arguments for selecting the research

methodology used in this study. Mixed methods were deemed sufficient because of

the robustness exhibited in other studies (Gil-Garcia, 2005; Venkatesh et al., 2013)

and the ability to capture the cultural uniqueness of Asian countries, particularly

Indonesia. Using this method, the study collected sufficient data and revealed subtle

information. Hence, the level of confidence for data completeness is considered to be

high. The study selected respondents carefully and thoroughly based on roles and

responsibilities in the Indonesian treasury as presented in Chapter 5.

94

THE INDONESIAN TREASURY E-GOVERNMENT

5.1 Overview

This chapter demonstrates why the Indonesian Treasury e-Government systems

could add more value to e-Government studies in developing countries. Some material

in this chapter is reused in the discussion presented in Chapter 8. This chapter begins

with Section 5.2 presenting the tasks of the Indonesian Treasury e-Government

systems. It follows with the antecedent of e-Government implementation in Section

5.3. Current e-Government systems development is presented in Section 3.4. Section

5.5 illustrates the evolution of ICT organisation. Finally, Section 5.6 presents how

decisions are made in regards to the Treasury’s ICT resources.

5.2 The Core of The Indonesian Public Financial Systems

The Indonesian Treasury e-Government system is considered to be the core of

the Indonesian public financial system as seen in Figure 5.1. Arrows in the figure

indicate data flow between related parties. According to the Indonesian constitution,

three state organs, i.e., the President, the House of Representatives and the Supreme

Audit Board, are involved in ensuring good governance in public financial

management from budgeting to reporting.

Once the budget is set, the Indonesian Treasury is involved from the beginning

to the end. Law No.1/2014 and Law No. 15/2014 mandated the Indonesian Treasury,

together with Supreme Audit Board, to develop a government accounting standard.

This standard is used for expenses, revenue and asset management when balancing the

finances. The cycle is ended with an annual government financial report for each

95

ministry. Using the same accounting standard, the Supreme Audit Board and the House

of Representatives oversee the ministries’ financial performance.

Law No. 1/2014 also mandated the Indonesian Treasury to manage the

government accounts in the Central Bank of Indonesia. In addition, to hasten

disbursement and collection of revenue, the Indonesian Treasury is allowed to have

more accounts in commercial banks. All accounts in commercial banks are

consolidated regularly to the government account in the Central Bank of Indonesia.

The Indonesian Treasury monitors and manages these accounts through its e-

Government systems.

The Indonesian Treasury e-Government systems must be able to communicate

with other systems within the Ministry of Finance. It begins with the budget data and

its variations. The data is received from the Director General (DG) of Budget. Budget

data is then used as a baseline for the limitation of expenses and for revenue targets.

The systems source revenue data from the DG of Taxation for tax revenue, the

DG of Customs and Excise for revenue from customs and excise, and the DG of Budget

and spending units for non-tax revenue.

Debt related data such as grants and loans is frequently communicated between

the Treasury and the DG of Debt Management. Asset management data from the DG

of Assets is exchanged regularly with the Indonesian Treasury systems to ensure

consistency of asset valuation and classification. The Indonesian Treasury e-

Government system should also exchange its data with the DG of Fiscal Balance to

warrant fair and just fiscal distribution throughout the country.

96

The Indonesian Treasury

e-Government Systems

Supreme Audit Board

- Gov. Financial Report- Accounting Standard

The President

- Gov. Financial Monitoring - Gov. Financial Report

House of Representatives

Gov. Financial Monitoring Gov. Financial Report

DG of Budget

- Budget Allocation- Non-Tax Revenue

DG of AssetsClassification, Appraisal of Fixed and Intangible

Assets

DG of Debt Mangement

Grant, Loan Management

DG of Fiscal Balance

Regional Fiscal Balance

DG of Taxes

Tax Revenue

DG of Customs and Excise

Customs and Excise Revenue

Commercial Banks

- Cash Disbursement- Collection of Revenue

Bank of Indonesia

Treasury Accounts

Government Personnel Agency (BKN)

- Personnel Salary and Allowance

Government Planning Agency (BAPPENAS)- National Programs

Monitoring

The Ministries

- Spending Units- Gov. Financial Report

Public Service Agencies (BLU)

Financial Management

Regional Governments

General and Special Allocation of Funds

The Ministry of Finance

State Organs

Government AgenciesCommercial Banks

(Private and State Owned Banks)

The Central Bank of Indonesia

Figure 5.1. The role of the Indonesian Treasury e-Government Systems

97

The Indonesian Treasury government systems should be able to communicate

with the e-Government systems of other government agencies. Examples of

government agencies using Indonesian Treasury data includes: the government

planning agency (BAPPENAS) monitoring budget outcomes in relation to the national

programs (Achmad, 2012); government personnel agency exchanging personnel

related data; and the Indonesian Treasury itself in order to support financial

management flexibility in the public service agencies (Susilo, n.d.); and, daily

transactions with more than twenty thousands spending units.

Commercial banks are involved in the process of disbursement of cash and

collection of government revenue. Law No. 17/2003 and Law No. 1/2004 mandated

the Indonesian Treasury to manage its cash through the Central Bank and through

commerical banks to ensure fiscal stability. The Indonesian Treasury receives daily

transactions for its accounts from commercial banks and the Central Bank of Indonesia

to its e-Government systems. Hence, the Indonesian Treasury is able to manage its

cash. In light of the above, it is seen that the Indonesian Treasury e-Government system

has a significant impact on the Indonesian government.

5.3 The Antecedent of e-Government Systems in the Indonesian Treasury

In supporting its role, the Indonesian Treasury has used ICT since the 1980s. The

technology used in the Indonesian Treasury has continued to evolve over time.

Initially, in the 1980s, the Indonesian Treasury used IBM mainframes to support its

business.

5.3.1 The Systems Evolution

In general, the evolution of e-Government systems in the Indonesian Treasury

can be divided into two phases: before and after the new millennium, as seen in Figure

98

5.2. Before year 2000, the government used mainframes due to several issues such as

financial, technical, and human resources. Paper based transactions dominated the

system design. The cost of ICT implementation in that era was enormous and, as a

result, the Indonesian Treasury maintained a single, centralised computing facility.

Figure 5.2. Systems Evolution in the Indonesian Treasury

All transactions in the Treasury Service Offices (TSO) nationwide at that time

were generated on paper using a typewriter. To ensure the accountability of the data,

the Indonesian Treasury generated multiple carbon copy duplicates for each

transaction. The paper-based transactions were regularly sent through the postal

system to the Information and Data Management Centre. Upon receiving the paper

transactions, the data was entered manually into the mainframe systems. This mode of

data transfer caused significant delay and errors resulting from data entry mistakes.

Later, facsimiles replaced the postal delivery of transactions to reduce the delay but

Typewriter Paper Only

FoxBASEFoxpro for Windows

Visual Foxpro and

MySQL

Mainframe (IBM DB2) Minicomputer (Oracle)

Year 2000

99

manual entry was still required resulting in no reduction in errors resulting from data

entry.

In the middle of the 1990s, as computers became less expensive, the Indonesian

Treasury rolled out computers to replace the typewriters. These computers ran

internally developed Indonesian Treasury systems that were based on FoxBASE

technology. From that time on, TSOs used computers to replace typewriters in

generating required treasury documents such as disbursement, bookkeeping, and

reporting. To ensure data validity, TSOs generated both an electronic copy and a hard

copy.

From the year 2000, minicomputers began replacing the mainframes and the use

of operating systems with graphical user interfaces (GUI), such as Microsoft

Windows, became prevalent. The Indonesian Treasury converted its FoxBASE

systems to FoxPro and Visual FoxPro for Windows. Most systems developed by the

Indonesian Treasury relied on the built-in database provided by FoxPro. As

requirements grew, the Indonesian Treasury introduced a client-server architecture in

the TSOs. The FoxPro database could not handle the needs properly and was replaced

with MySQL.

As the Indonesian Treasury shifted from a centralized to a distributed system,

data communication became essential.

5.3.2 Data Communication

In the mainframe era, data communication to connect all offices was

unnecessary. Due to a new millennium compliance issue, the Indonesian Treasury

shifted its computer technology to the IBM minicomputer. The project was called

downsizing.

100

In the downsizing project, the Indonesian Treasury started using personal

computers (PCs) with Local Area Network (LAN) capability. Paper was the only

accepted document for Treasury related transactions, PCs were then used to generate

electronic data to satisfy paper-based transactions. Due to lack of telecommunication

infrastructure, generated data from all offices was sent regularly using diskette(s) and

paper to the Information and Data Management Centre.

A year after the downsizing project, the Indonesian Treasury started to use the

Internet as a means of transferring data from the TSOs. However, not all TSOs could

use the Internet because of limitations in the available telecommunication

infrastructure. Figure 5.3 shows that at the beginning of the new millennium, three

indicators, i.e., telephone lines, Internet users and fixed broadband subscribers, were

very low.

Figure 5.3. Telephone, Internet and Fixed Broadband subscribers in Indonesia

per 100 people (The World Bank, 2014)

0

2

4

6

8

10

12

14

16

18

19

99

20

00

20

01

20

02

20

03

20

04

20

05

20

06

20

07

20

08

20

09

20

10

20

11

20

12

Telephone lines (per 100people)

Internet users (per 100people)

Fixed broadband Internetsubscribers (per 100people)

101

The use of data communication in the Indonesian Treasury evolved over time.

A few years after the enactment of the Indonesian financial laws, i.e., Law No.

17/2003, Law No. 1/2004 and Law No. 15/2004, the Indonesian Treasury successfully

installed its first Wide Area Network (WAN). The Indonesian Treasury referred to this

WAN as the Intranet.

The Intranet connects all offices in the Indonesian Treasury by using

Multiprotocol Label Switching (MPLS) technology. The network was developed with

assistance from one of the largest Indonesian telecommunication companies. The

presence of the Intranet improved accessibility, availability and reliability of data

communication. Although the Intranet became available 24 hours a day, systems only

used it for electronic data transfer, data transfer monitoring and other non-core

transactions. There were no online and real-time transaction systems running over the

network. The nature of system applications in all offices remained similar to those

used before the deployment of the Intranet.

Realising what data communication could deliver, the Indonesian Treasury

decided to shift from distributed systems to centralised one. A new project initiative

came into place in 2004 with assistance from the World Bank after the enactment of

the Indonesian public financial laws.

The project is called Sistem Perbendaharaan dan Anggaran Negara (SPAN) or

state budget and treasury systems. The SPAN adopted a commercial off-the-shelf

solution, instead of developing the system from scratch. As of the end of 2014, the

project had not been fully implemented. Thus, when this study concluded its data

collection, the SPAN had not been fully used for one budget cycle in the piloting

102

offices. Full rollout of the SPAN to all offices was announced in 2015. Hence, the

SPAN was excluded from this study.

5.4 E-Government Systems in the Indonesian Treasury

The Indonesian Treasury uses Microsoft FoxPro as the main software to develop

their systems. The Indonesian Treasury follows regulation number 351/KMK.01/2011

from the Ministry of Finance regarding the Policy and Standard of Information

Systems Development Cycle within the Ministry of Finance. This regulation dictates

that information systems development should adhere to a Software Development Life

Cycle (SDLC).

5.4.1 Software Development Life Cycle

SDLC refers to the approach taken in the development of software (Langer,

2008, 2012). The SDLC framework in the Ministry of Finance consists of requirement

analysis, systems design, implementation, testing, deployment and evaluation as seen

in Figure 5.4.

There are many SDLC models including the traditional Waterfall model,

Iterative, Spiral, V-model, Big Bang, Agile, Rapid Application Development and

Prototyping (Langer, 2012). Although the Indonesian Treasury defines a generic

framework for the SDLC in developing their systems, there is no prescription on any

particular SDLC model to use. Regardless, the Indonesian Treasury still managed to

develop the e-Government system to support treasury functions.

103

Figure 5.4. SDLC Framework

5.4.2 The Systems

The systems were developed with the multi-level hierarchy of the Indonesian

government organisations in mind. According to Law No 1/2004, the treasury

functions in Indonesia are applied to the ministries and the Treasurer as seen in Figure

5.5. Each level in a ministry is paired with a unit of the Treasurer. The e-Government

systems reflect this structure being run in stages from the lowest to the highest level.

The Indonesian Treasury developed all related systems to satisfy the treasury

functions. It is compulsory for all Indonesian government agencies to use the systems.

The Indonesian Treasury argues that uniformity of data is the main reason for

distributing the systems to all spending units.

Analyse

Design

Develop

Testing

Deploy

Evaluate

104

Spending Units-The Ministries

The Indonesian Treasurer

The TreasurerThe Ministry

Ministry

Echelon 1

Headquarters

Regional

Spending Units

TROs

TSOs

Scope of the study

Figure 5.5. Separation of Treasury Function

The Indonesian Treasury’s total system is distributed in nature, with each

individual, installed system having its own database. A brief systems configuration

can be seen in Appendix H.

The business model of the Indonesian Treasury forces all units to transfer the

output of one system to other systems. Therefore, each system needs to communicate

with other systems as shown in Appendix H. The result of each system is then

105

transferred in the form of hardcopy, electronic copy, or both. All systems are equipped

with functionality to satisfy such processes. Although sufficient training and

documents for each system are provided, the Indonesian Treasury still experiences

data discrepancy.

The Indonesian Treasury invested in business intelligence applications (Watson

& Wixom, 2007) as its decision support system (Sol, 1987) to generate more value

from its scattered systems. Negash (2004) argues that business intelligence could

increase the value of information by combining operational data and analytical tools.

Care is needed, however, as business intelligence can generate insufficient information

due to processing of unsound data.

5.5 ICT Organisation Evolution

The Indonesian Treasury has reorganised its ICT department four times since

2004 as seen in Figure 5.6. Initially, a conservative approach was used (Laudon &

Laudon, 2011). ICT was part of the accounting directorate. The ICT department was

separated from the accounting directorate in 2006 to form the Directorate of Treasury

Systems (DTS). The Directorate of Accounting and reporting was responsible for

accounting related functions, while the DTS was designed for information systems

development.

Although the DTS was originally designed for information systems related

development, its actual role was not clearly stated in the Ministry of Finance (MoF)

regulation number 466/2006. This resulted in multiple interpretations when the DTS

was established. One of the six Deputy Directors, specifically, the Deputy Director for

treasury systems development, was designated to be a business analyst that would

implement treasury related regulations into the system. Rather than playing this role

106

Figure 5.6. ICT Organisation Evolution

Directorate of Information and Accounting

Directorate of Treasury Systems

Directorate of Accounting and Reporting


Directorate of Treasury Transformation


Directorate of Treasury Information Systems

and Technology

2004 2006 2008 2014

107

as business analyst, the Deputy Director of treasury system development became a

division that worked on harmonising treasury regulations.

To hasten e-Governement systems development, the Indonesian Treasury

reorganised its ICT department through Ministry of Finance regulation number

100/2008. In addition to the Directorate of Treasury Systems, a Directorate of Treasury

Transformation (DTT) was formed. DTT was responsible for the future of the

Indonesian Treasury e-Government systems, leaving the DTS to be responsible for

ensuring current systems ran properly with minimum alteration.

Subsequently, the role of the DTS was redefined. The need for harmonising

treasury regulations and refining busisness processes was paramount. In the DTS, a

division of treasury regulations and development was introduced to replace the

treasury systems development division. Two ICT related divisions, namely system

application development and ICT support, were retained. These two ICT divisions

were assigned to work closely with DTT.

Additionally, the DTT took on the role of ensuring that the new initiative of

adopting commercial-of-the-shelf (COTS) solutions in the sistem perbedaharaan dan

anggaran negara (SPAN) project ran smoothly. The DTT is equipped with two

business process analyst divisions and two ICT related divisions. Hence, the DTT and

the DTS have similar roles. Although the DTT and the DTS were assigned to work

closely and collaboratively, confusion between them persists.

A few months after this study completed its data collection, the Ministry of

Finance introduced, as a result of adopting balanced scorecard (BSC), regulation

number 206/2014 to reorganise the Indonesian Treasury. The SPAN project was

considered nearly complete and, as a result, the DTT and the DTS were reformed.

108

However, this study could not find any documentation to support this ICT organisation

evolution. Hence, leadership could play a significant role in the evolution. Taking

everything into account, in developing its e-Government systems, the Indonesian

Treasury seems to have encountered trials and errors in its ICT organisational

evolution.

5.6 ICT Resources

ICT resources in the Indonesian Treasury can simply be divided into hardware,

software and human resources. The Indonesian Treasury has invested in all these

resources. The decision as to what kind of hardware and software should be purchased

was taken conservatively by asking its enterprise entities.

5.6.1 Hardware and Software

The Indonesian Treasury relies on various kinds of computers. Among those,

around 30 percent are considered to be obsolete, as seen in Figure 5.7. Yu et al. (2010)

suggest that the average computer lifespan in the world varies from 3.8 to 5.75 years

old. Hence, the age of the computers used by the Indonesian Treasury could result in

insufficient support for their maintenance.

109

Figure 5.7. Total Computer Assets as of December 2013

Most computers are running a Microsoft Windows operating system. However,

this study could not get any information regarding how many computers are using the

latest version of Windows, as depicted in Figure 5.8. In the latest computers and assets

survey, detailed computer configuration was excluded. The latest detailed

configuration data is only available up to the end of 2007.

Figure 5.8. Operating Systems Used as of 2008

189

3618

8499

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

Purchased before 2000 Purchased between 2000 -2007

Purchased after 2007

COMPUTER ASSETS

15

15

30

89

521

3222

Windows ME

Windows Longhorn

Windows 98

Windows Server 2003

Windows 2000

Windows XP

0 500 1000 1500 2000 2500 3000 3500

110

Data from 2007 probably does not reflect current computer configurations.

However, some useful information can still be analysed. Most computers were using

Windows XP. This study confirmed during its data collection that the majority of its

computers are still using Windows XP. However, Microsoft has announced that they

are no longer supporting Windows XP with any updates from April 2014 (Microsoft,

2014b). As a result of not getting any update for Microsoft XP, the Indonesian

Treasury may be prone to security threats.

Although most of the e-Government systems were developed using Microsoft

FoxPro, they are run with various databases. FoxPro has the ability to connect to

different databases, including its built-in database. As presented in Appendix H, most

systems used FoxPro DBF. A small number of systems use MySQL or another Oracle

database. The databases are installed on several server platforms.

The Indonesian Treasury adopted various operating systems for its server

platforms. The majority of its servers used Microsoft Windows Server 2003, as shown

in Figure 5.8. The Indonesian Treasury may not be aware of recent announcements

that Microsoft will no longer support Windows 2003 servers from mid-2015

(Microsoft, 2014a). In addition to the Windows server platform, the Indonesian

Treasury also used various Unix based servers. Variants of Unix platforms evident in

the Indonesian Treasury data centre include Solaris, IBM AIX, and Linux.

Hardware and software vendors often develop their own certification programs

for qualified engineers (Adelman, 2000; Schlichting & Mason, 2004). Thus, a pool of

ICT certified human resources are needed to ensure all resources continue running

properly.

111

5.6.2 Human Resources

The Indonesian Treasury recruits its employees by following a general rule for

government employee recruitment in Indonesia. Recruitment is only done at an entry

level through its ministry. The process can take around a year from defining the needs

until awarding the applicants. Generally, less than 10 percent of applicants recruited

into the Indonesian Treasury are for ICT related fields.

In addition to formal education, the skill of ICT related staff needs to be proven

by obtaining professional certification (Parnas, 1990; Tondeur et al., 2007). In 2014,

across all staff in the Indonesian Treasury, there were only around 100 employees

holding bachelor degrees in ICT related fields out of a total of 4617 employees holding

some form of Diploma and/or Bachelor Degrees. This is depicted in Figure 5.9.

Further, in two directorates with responsibility for ICT, there were only around 30

employees with an ICT background. Also, the number of qualified engineers was

considered scarce with less than ten people holding Cisco certification.

Figure 5.9. Education Background of all staff in the Indonesian Treasury

2604

4617

816

Education Background

Up to High School Diploma and Bachelor Postgraduate

112


As the back-office (Dawes, 2008) for public financial management systems, the

Indonesian Treasury e-Government systems play an important role in Indonesia. The

systems impact other systems in other organisations, such as state organs, various

government agencies, and banks. Thus, the Indonesian Treasury e-Government

systems require more than just a conservative approach. It requires an instrument that

can be used not only to develop the systems effectively, but also to communicate with

large numbers of parties.

In addition to systems development, this chapter also reflects that the Indonesian

Treasury experienced complications in managing its organisation evolution and ICT

resources. Enterprise Architecture (EA) could be used to manage such complexity

(Op't Land et al., 2009; Townson, 2008; Zachman, 1997). In order to validate whether

EA could have an effect on the Indonesian Treasury e-Government system

performance, this study uses proposed research model in Chapter 3 and mixed method.

The presentation of the collected data begins in the next chapter with the findings of

the qualitative analysis.

113

QUALITATIVE ANALYSIS

6.1 Overview

The chapter presents the results from the qualitative analysis of this study. The

phases of qualitative analysis are presented in Section 6.2. Respondent profiles are

presented, ensuring anonymity, in Section 6.3. To ensure rigor in the findings, the

results of the semi-structured interviews were reviewed using content analysis,

presented in Section 6.4, and thematic analysis, presented in Section 6.5. The findings

from the interviews were confirmed using document analysis and observation as

discussed in Sections 6.6 and 6.7.

6.2 Phases of Analysis

The study followed Yin (2011) in analysing qualitative data. He suggested a five-

phase cycle in analysing qualitative data as seen in Figure 6.1.

First, it begins with compiling the data into a database. The study transcribed all

interviews and each transcript was sent to the interviewee for comment and to control

bias. The confirmed transcripts were archived in files in the Portable Document Format

(PDF). Documentary evidence collected was also kept as PDF files.

The second phase was the process of extracting data from the documents in the

database and coding the data accordingly. This study started by classifying the two

main sources of data: interview transcripts and documentary evidence.

The third phase, reassembling, is used to obtain insight into any patterns in the

data. Before concluding the analysis, the researcher needs to interpret the data, this is

the fourth phase. This study used computer software called NVivo version 10 to assist

114

analysing the data. Finally, in phase five, the results of the analysis are presented after

describing the profiles of the respondents who were involved in the study.

Figure 6.1. Five-phase cycle (Yin, 2011, p. 178)

6.3 Respondent Profiles

This study identified twenty feasible interviewees from the Treasury staff.

Feasibility was based on the current position held and previous contributions made to

the e-Government systems development. To maintain anonymity, only the role of the

interviewee and the date of the interview are seen in Seventy five per cent of the people

approached responded positively to the request for interview, as seen in Table 6.2. The

remaining twenty five per cent of people approached did not respond even though

several gentle reminders were sent.

4. Interpret Data

3. Reassemble Data

2. Disassemble Data

1. Compile Database

5. Conclude

115

The majority of respondents came from Echelon Three. Interviewees for this

rank were situated only in the headquarters. Responses from this rank in the TROs and

TSOs were collected using questionnaires (described in the next chapter).

Table 6.1. All interviewees were given a random respondent ID.

Seventy five per cent of the people approached responded positively to the

request for interview, as seen in Table 6.2. The remaining twenty five per cent of

people approached did not respond even though several gentle reminders were sent.

The majority of respondents came from Echelon Three. Interviewees for this

rank were situated only in the headquarters. Responses from this rank in the TROs and

TSOs were collected using questionnaires (described in the next chapter).

Table 6.1. Interviewee Role and Interview Date

Ech Role Date

3 Deputy Director of Database Management and ICT Support 26/11/2013

2 Head of Treasury Regional Office of Bali 5/12/2013

3 Deputy Director of System Application Transformation 9/12/2013

2 Director of Treasury Systems 12/12/2013

3 Head of the Indonesian Treasury Human Resources 12/12/2013

3 Deputy Director of System Application Development 13/12/2013

3 Deputy Director of ICT Transformation 17/12/2013

2 Head of Treasury Regional Office of West Java 23/12/2013

2 Head of ICT Centre for the Ministry of Finance 9/01/2014

3 Head of the Indonesian Treasury General Affair 9/01/2014

2 Head of Treasury Regional Office of the Islands of Riau 10/01/2014

3 Deputy Director of Transformation Support 10/01/2014

1 Special Staff for ICT 13/01/2014

2 Director of Treasury Transformation 13/01/2014

1 Secretary General of the Ministry of Finance 17/01/2014

116

Table 6.2. Response to Interview Request and Respondent ID based on ranks

Level Able Unable PCT per rank

Echelon 1 2 2 50%

Echelon 2 6 2 75%

Echelon 3 7 1 87.5%

Total 15 5

PCT 75% 25%

At the time the interviews were conducted, there was only one respondent aged

less than forty years old. The main age groups were between 41 and 45 years old and

more than 56 years old, as reflected in Table 6.3.

Table 6.3. Age distributions

Age Total PCT

Less than 40 1 7%

41 – 45 5 33%

46 – 50 2 13%

51 – 55 2 13%

More than 56 5 33%

This study decided to conduct the majority of interviews in the Treasury

headquarters, as depicted in Table 6.4. The main reason for this is that e-Government

system development and policy formulation are mainly made in the headquarters.

However, respondents from regional offices, such as the head of regional offices who

were not giving their responses via the questionnaires, were also invited. The purpose

of this invitation was to understand the regional offices’ view and contribution to the

e-Government systems development.

117

Table 6.4. Respondent’s Location

Location Total PCT

Headquarters 12 80%

Regional Office 3 20%

The ratio of respondents who are heavily involved in the systems development

and those who have more general roles are relatively equal as seen in Table 6.5.

Table 6.5. Respondent’s Role

Job Roles Total PCT

General 7 47%

Systems Development 8 53%

All respondents had at least a bachelor degree and all but one had completed

some form of postgraduate education, as seen in Table 6.6.

Table 6.6. Educational background

Level Total PCT

PhD 2 13%

Master 12 80%

Bachelor 1 7%

Although there were eight interviewees involved in systems development, only

two respondents had ICT related degrees. Most respondents came from economics and

public management backgrounds, as seen in Table 6.7.

Table 6.7. Respondent’s educational background

Field Total PCT

Economics 6 40%

ICT 2 13%

Management 7 47%

118

6.4 Content Analysis

Content analysis refers to “a research technique for making replicable and valid

inferences from texts (or other meaningful matter) to the contexts of their use”

(Krippendorff, 2004, p. 18). According to Elo and Kyngäs (2008) content analysis

begins with data preparation, organising the data using open coding, and reporting the

results through a model. NVivo was used to assist this process in this study.

This study began with importing the electronic copies of the interview

transcripts. Data were organised into groups of folders. Word frequency queries were

made on the documents. The study conducted queries across four groupings of

respondents: all respondents, Echelon 1, Echelon 2 and Echelon 3. The queries used

the following criteria: exact matches, display the 200 most frequent words and words

had to contain at least five letters.

All words appearing in the results were thoroughly analysed. Some words in the

Indonesian language such as untuk, sehingga, belum, seperti, and dapat were omitted

from the query results. The English equivalents of these words represent grammatical

articles such as conjunctions and are considered to be expressions irrelevant to the

study. Some words that can be used interchangeably and have similar meanings were

combined into one word. The study used the English “information” for both informasi

(information) and laporan (report). Although those two words have different semantic

meaning, interviewees used these words in similar contexts.

This study presents the ratio of the top fifteen words identified. The ratio of the

frequency of each word and the total frequency of the fifteen most frequently used

words is presented as a percentage. The results for the overall grouping and the

119

individual, Echelon based groups of respondents are presented in the subsequent

sections.

6.4.1 All respondents

The fifteen most frequently used words by all respondents can be seen in Figure

6.2. The data shows that all respondents had big concerns regarding the “systems”,

“information” and “hardware” with the totals of 21%, 15% and 13% respectively. The

least frequent words used by the respondents were “integration”, “operational” and

“alignment”.

Figure 6.2. Most Frequently Used Words by All Respondents

6.4.2 Echelon 1 (2 Respondents)

The study further analysed the data for each of the Echelon-based respondent

groupings. In comparison with all respondents, the respondents who were in Echelon

1 had different concerns reflected by their most frequently used words. Rather than

speaking about the “systems”, they mentioned “performance”, “operational” and

“architecture” more frequently with 23%, 14% and 11% as seen in Figure 6.3.

Information15%

Systems21%

Hardware13%

Process10%

Software10%

Business8%

Investment7%

Architecture3%

Network3%

Database3%

Operational2%

Performance2% Integration

2%

Alignment1%All

120

“Integration” was the only word in common with the three least frequently used words

by all respondents. The other two words were “network” and “business”.

Figure 6.3. Most Frequently Used Words by Echelon 1


The Echelon 2 respondents showed different interests compared with the

Echelon 1 respondents. The Echelon 2 respondents more frequently spoke on strategic

and tactical topics such as “systems”, “information”, and “process”, as seen in Figure

6.4. They were less attentive to topics of “alignment”, “integration”, and

“performance”.

Automation2%

Budget5% Information

5%

Systems5%

Hardware11%

Process5%

Business2%

Investment11%Architecture

11%

Network2%

Operational14%

Performance23%

Integration2%

Alignment2%

Echelon 1

121



In the interview, Echelon 3 respondents were more concerned with tactical and

pragmatic topics such as “systems”, “hardware” and “information”, as reflected in

Figure 6.5. These respondents were less concerned for “performance”, “standard” and

“accountability”. These three words were the least frequently used by the Echelon 3

respondents.


Automation2%

Information15%

Systems20%

Hardware8%

Process12%Software

11%

Business10%

Investment5%

Architecture3%

Network3%

Database5%

Operational3%

Performance2%

Integration2% Alignment

2%Echelon 2

Governance3%

Information13%

Standard2%

Synergy4%

Systems20%

Hardware15%Process

7%

Software9%

Business5%

Investment11%

Architecture5%

Network3%

Performance2%

Integration2%

Echelon 3

122

6.4.5 Comparison

During the interviews it was evident that there were different interests between

the groups of respondents. This can be inferred from the comparison of the fifteen most

frequently used words as seen in Table 6.8. Echelon 1 and Echelon 2 respondents used

words that reflected strategic views such as “performance”, “automation”, and

“budget”. Echelon 2 respondents also expressed interest in technical perspectives as

indicated through use of words “software” and “data” (database), although the use of

these words was usually in a tactical context. This can be seen from the following

responses from Respondent 6:

“The Indonesian Treasury e-Government is in the transition of shifting from

distributed databases to an integrated one.”

Respondent 4 highlighted the following message:

“Due to multiple databases used for different purposes, the needs for centralised

common expenses such as a monthly payment for utilities and salary cannot be

simplified from thousands of transactions to only one transaction.”

Table 6.8. Comparison of Most Frequently Used Words by All Respondents

All Echelon 1 Echelon 2 Echelon 3

Automation - 2.17% 2.08% -

Budget - 4.35% - -

Governance - - - 2.58%

Information 14.29% 4.35% 14.58% 13.30%

Standard - - - 1.72%

Synergy - - - 3.86%

Systems 20.29% 4.35% 19.58% 20.17%

Hardware 13.04% 10.87% 7.50% 14.59%

Process 9.73% 4.35% 11.67% 6.44%

Software 9.52% - 10.83% 8.58%

Business 8.28% 2.17% 10.42% 5.15%

123

Investment 6.63% 10.87% 4.58% 10.73%

Architecture 3.11% 10.87% 3.33% 4.72%

Network 3.11% 2.17% 2.50% 3.43%

Database 2.90% - 5.42% -

Operational 2.07% 13.04% 2.50% -

Performance 2.07% 21.74% 1.67% 1.72%

Integration 1.86% 2.17% 1.67% 1.72%

Alignment 1.45% 2.17% 1.67% -

All in all, it is seen that the three groups have different concerns regarding the

Indonesian Treasury e-Government. The Echelon 1 respondents were more concerned

with the performance of the systems whereas the other two groups had a similar level

of concern for systems and information as seen in Figure 6.6. This indicates that each

level demonstrated different interests. It is consistent with the EA framework proposed

by Zachman (1997). He drew four different levels of roles for those involved in IS

development within an enterprise. Further analysis for each theme in the interview

responses are provided in Section 6.5.

124

Figure 6.6. Radar View of Frequent Words Used

6.5 Thematic Analysis

Similar to content analysis, the use of thematic analysis as an approach to

analysing qualitative data is very common (Vaismoradi, Turunen, & Bondas, 2013).

Thematic analysis is defined as “a method for identifying, analysing and reporting

patterns (themes) within data” (Braun & Clarke, 2006, p. 79). According to Braun and

Clarke (2006) thematic analysis begins with becoming familiar with the data by

reading it thoroughly. It is then followed by “generating initial codes, searching for

themes, reviewing themes, defining and naming themes, and producing the report”.

This study followed Bazeley and Jackson (2013) in its conduct of the thematic

analysis of its data using NVivo 10. Codes were made according to the research

framework introduced earlier in Section 4.3. Searching for themes was conducted

through a text search menu by giving pertinent key words. The results were then

reviewed. The themes were named as nodes: Architecture, Alignment, Information,

Performance, and Resources in NVivo (Bazeley & Jackson, 2013). Each of the nodes

represents a theme in the framework. However, the Resources node was used to

0.00%

5.00%

10.00%

15.00%

20.00%

25.00%Automation

BudgetGovernance

Information

Standard

Synergy

Systems

Hardware

ProcessSoftwareBusiness

Investment

Architecture

Network

Database

Operational

Performance

IntegrationAlignment

All

Echelon 1

Echelon 2

Echelon 3

125

represent both the resource portfolio optimisation and resource complementarity

themes because they are relatively similar. The results are presented in the following

sections.

6.5.1 Enterprise Architecture

This study could not find any artefact showing Enterprise Architecture (EA) or

similar being present in the Indonesian Treasury. All interviewees were asked about

the presence of ICT related architectural documents such as EA. The common

response was:

“To date such documents are not available” (Respondent 1).

Respondent 1 further argued “In my view, most people in the Indonesian

Treasury view ICT like other peripherals such as cars. We only consider using cars to

take us to our destination. Conversely, ICT requires proper planning and governance.

As such, I would say there is no reference as to where ICT should be developed.”

Similar to responses from Respondent 1, Respondent 4 argued:

“We never had any ICT related architecture documents. We do not know what

we are going to develop within the next five years. Even in the technical levels,

we did not have proper documentation as to how ICT is developed. Technical

documentation for the e-Government systems are also absent. Thus, e-

Government systems development is controlled by only a few people.”

Another interviewee, Respondent 3 corroborated with the following:

“The direction of existing e-Government systems remains unclear as to how they

should be developed because there is no architecture as its guidance.”

A different interviewee, Respondent 12 discussed:

“Since there is no architectural documentation both at the Ministry of Finance

level and the Indonesian Treasury, as an e-Government systems development

framework, we did not know how ICT will be utilised in the future. Thus, to some

extent I think our investment in ICT is not efficient”.

Another interviewee, Respondent 10 pointed out:

126

“I heard there was an initiative to create an ICT blueprint. Yet I had never read

such a document. Thus, currently systems are developed upon request from

users. Missing architectural documentation causes each directorate to only view

their needs. As a result they demonstrate lack of respect to others.”

In addition to responses from interviewees residing in headquarters, interviewees

from regional offices also confirmed that there was no architecture documentation.

Some interviewees referred to them as blueprints, such as Respondent 5 stating:

“E-Government systems require a blueprint so that it can be used as a point of

reference for everyone involved.”

Another interviewee, Respondent 7, mentioned:

“To the best of my knowledge, there is no complete e-Government design to

come from Jakarta (headquarters). Thus, I do not know of any documentation as

a guideline for us in the regional offices.”

However, one of the fifteen interviewees did not give a clear response.

Respondent 11 said:

“I heard about it. However, the Indonesian Treasury does not have lots of people

who understood what it would bring. I believe that at the ministerial level

(Ministry of Finance), we have such documents. A government unit under the

ministry such as the Indonesian Treasury was expected to develop one too.”

Despite the absence of architectural documents for the Indonesian Treasury e-

Government systems, the Indonesian Treasury manages to run their businesses by

relying on other instruments. Using balance scorecard, the Indonesian Treasury

defined key performance indicators (KPI) for its offices nationwide. Similarly, for

offices in the field of e-Government systems, the Indonesian Treasury is using KPIs

as stated by Respondent 13:

“Although to date an ICT blueprint is not available, all offices in the Indonesian

Treasury, including ICT units, are given KPIs. These KPIs should be

accomplished annually.”

127

Notwithstanding responses regarding lack of architectural documentation in the

Indonesian Treasury e-Government systems, there is a growing awareness of the need

for architecture documents. This is reflected in responses such as:

“My staff is starting to develop architecture documentation for ICT

infrastructure, data, and others. I am not sure whether or not they have sufficient

knowledge about it because they are technical people” (Respondent 3).

Another respondent, Respondent 5 stated:

“We needed a blueprint as guidelines. To date, systems application development

is heavily reliant on individuals, which is not acceptable. We are heading in that

direction. We understand the needs of proper design to make the most of our ICT

investment”.

Respondent 4 also stated that:

“Unlike before 2010, we now know how our core businesses should be run, such

as cash management, and government accounting in the Indonesian context. We

needed a comprehensive plan and how to achieve it. That includes managing the

impact of such a plan. I believe the plan will have a significant impact on our

organisation.”

It can be inferred that the Indonesian Treasury are not referring to any architectural

document in developing their e-Government systems.

6.5.2 Organisational Alignment

The study asked its respondents whether they shared a common understanding

of the strategic goals of e-Government and what they contributed towards achieving

these goals. Different responses were received during the interviews. Most

respondents replied by saying that common understanding did not exist. They had a

tendency for answering the question by saying the alignment was insignificant or even

that no alignment existed, such as:

“Alignment is a long way off” (Respondent 4).

Respondent 4 further used his experience by stating:

128

“when I was involved in designing complex systems, there was nearly no

significant direction as to how the systems should be developed. Most initiatives

took a bottom up approach rather than top down. ICT should be treated as a

strategic project and sufficient direction should come from the top”.

Another interviewee (Respondent 3) confirmed by saying:

“In the existing e-Government systems, we cannot estimate government expenses

per month or semester. We are not sure whether or not the provision of funds is

sufficient. Therefore, I would say the existing systems do not fully align with the

Indonesian Treasury strategic mission.”

Some respondents mentioned holistic views, which could lead to common

understanding, were not available, such as:

“Common understanding of the Indonesian Treasury vision and mission can be

seen from the e-Government systems development, which is considered to be

partial. Users only see their roles and responsibilities by overlooking others.

The need for a common framework to bring them to the same viewpoint is

needed” (Respondent 10).

In addition, another interviewee, Respondent 15 said:

“up until now we have developed the systems based on partial needs. They have

not been combined into one integrated system”.

Respondent 14 supported the statements by saying

“When the existing e-Government systems were launched, it was found that they

did not fully fit with the business process requirements”.

Another interviewee gave an example of such a situation:

“When we deployed e-Government systems, we found out that they were not

referring to the business process we designed for, the treasurer bookkeeping

systems report” (Respondent 8).

This situation might come from leadership issues as stated by Respondent 12:

“The e-Government systems development cannot be hastened or improved due

to lack of sufficient support from the policy makers.”

However, not all respondents gave the same response. One respondent,

Respondent 6 stated that:

129

“To date, the e-Government systems have aligned with the Indonesian Treasury

strategic mission. However, I believe there is still room for improvement. I was

convinced that we are walking on the right path. The systems will be

continuously developed”.

Another interviewee, Respondent 5 provided similar response saying:

“ICT has been utilised in alignment with the business process. ICT has helped

the service performance. Yet the current systems are not fully optimised because

some improvements are still needed”.

According to responses from the interviewees, it can be inferred that the

Indonesian Treasury are still struggling in aligning with their organisation entities.

6.5.3 Information Availability

The Indonesian Treasury seems to be struggling in gathering reliable information

in a timely manner. This can be seen in some of the responses given by interviewees

during the data collection. For example, Respondent 11 stated:

“Reports presented to high level officials are sometimes delayed by about a

month. There may be some problem in the financial process itself or the

reporting process. The delay causes a loss in value derived from the reports for

the high level officials.”

Likewise, one of the respondents from Echelon 1, Respondent 2 stated:

“There were circumstances where we needed additional time in making some

decisions caused by the delays in generating the financial report”.

Similarly, Respondent 6 stated:

“To date the data quality is considered low. The quality is getting better,

however. We are aiming for data perfection.”

Other interviewees supported such statements by saying:

“Lack of discipline in managing data results in a lot of time being required for

data cleansing. Therefore, the Indonesian Treasury cannot be fully responsive

in providing reliable information” (Respondent 1);

and

130

“We were too permissive to the ministries and spending units as budget users.

This resulted in data variation” (Respondent 3).

The Indonesian Treasury has some issues regarding how the e-Government

systems are running within its offices as well. First, the way reports are made, as stated

by some respondents, such as:

“The Treasury Service Offices send two sets of data to the headquarters i.e. data

transaction and general ledger (GL). Data transaction is used to generate

financial managerial reports called BUKU MERAH in Indonesian. Data in the

general ledger is used to generate accounting based financial reports. This

results in two different reports. In short, I would be hesitant to call these valid

reports” (Respondent 4).

Another interviewee said:

“I found some discrepancies when comparing data from the Treasury Service

Offices and the headquarters” (Respondent 12).

In addition, extra effort was needed involving humans in generating the reports.

Human error is considered to be one of the causes of system failures (Brown &

Patterson, 2001; Dix et al., 1997). Therefore, this could affect the quality of

information. Respondent 8 stated:

“Current e-Government systems can generate invaluable reports such as:

Actual Government Budget Report, Cash Flow Report, and Government

Financial Report. They are not automatically generated by the systems. The

systems are used to make data collection easy but not to generate reports. The

reports are made manually.”

Second, there are complications within the e-Government systems, as stated by

Respondent 12:

“Most supervisors have backdoor access straight through to the database. They

can manipulate the data without leaving any trace. This harms both the systems

and the data transactions.”

Another interviewee, Respondent 14, supported this statement by saying:

“Even though it came from one transaction, data discrepancies are still

common.”

131

Third, e-Government systems were developed reflecting each individual

business. Regarding this, Respondent 2 said:

“The Indonesian Treasury business units separately managed their own data.

This led to multiple stages of data consolidation. However, we are in the process

of integrating them all in the SPAN.”

Other respondents made similar statements in this regard:

“Causing extra effort and processes to bring together the data into valuable

information in the form of reports” (Respondent 11); and

“What we sent to Jakarta was not immediately updated in the systems causing

some time lag for the TSOs to run their businesses” (Respondent 7).

Thus, time wise, the Indonesia Treasury has to be tolerant in receiving data and

generating reports from its offices (Respondent 6; Respondent 10)

Fourth, unclear governance as to how the data should be managed also

emmerged in the study as in the following statement:

“Governance as to when the information can be accessed by the Treasury staff

and who are entitled to access such information is missing” (Respondent 11).

Further, Respondent 11 said:

“Not all leaders in the Indonesian Treasury are aware of what ICT can deliver,

such as being able to create their own customised reports directly from the

systems. As of now, there are no decision support systems used by the leaders

that would aggregate all transactions nationwide.”

In contrast to most responses, Respondent 5 stated:

“In certain cases, the Indonesian Treasury e-Government systems have been

able to provide adequate information. This information can be directly used by

decisions makers and other units.”

Also, from Respondent 9:

“Advisors to the President rely on the reports produced by the Indonesian

Treasury using the systems and resources currently available to the Treasury.”

132

6.5.4 Resource Portfolio Optimisation

Investment in ICT resources in the Indonesian Treasury is not considered fully

satisfactory. This can be seen from most respondents stating that unnecessary

investments were still occasionally present. The Indonesian Treasury tried to minimise

further problems by referring to the data on obsolete resources according to a standard

for economic life and issued requests to vertical offices nationwide in deciding where

to invest in ICT (Respondent 15). Thus, the Indonesian Treasury headquarters

regularly asks vertical offices (TROs and TSOs) to send their needs requests

(Respondent 6; Respondent 15). The needs are then used to formulate technical

specifications with the ICT Directorates: Directorate of Treasury Systems (DTS) and

Directorate of Treasury Transformation (DTT). Thus, the Indonesian Treasury is not

referring to an ICT plan (Respondent 11).

Such a plan has never been formed in any ICT blueprint, architecture or

framework (Respondent 12). Respondent 12 further argued that there was no clear

direction as to how and where to invest in ICT because ICT was still viewed as tools.

This leads to a lack of awareness for proper planning (Respondent 1; Respondent 12).

Missing a proper design for ICT resulted in insufficient program budget proposals

(Respondent 1). The budget constraints were also caused by lack of awareness, from

the leaders, of the importance of having reliable e-Government systems (Respondent

3). Thus, there is lack of support for driving suitable investment in ICT resources.

Furthermore, another respondent, Respondent 5 stated that:

“we might not need to have high specification computers to run the business”.

Similarly, another respondent, Respondent 7 mentioned:

133

“There was no control as to what computer specifications should be purchased

and how the PCs should be used. We were inclined to purchase the most

sophisticated computers without knowing the real needs”.

Therefore, it is understandable why Respondent 4 said that ICT spending was

not generating additional value to the Indonesian Treasury. Although the Indonesian

Treasury had a constrained budget (Respondent 3), another interviewee stated:

“When I was working in the TRO, I found that on the one hand some TSOs

received new computers when there was no need for additional computers, on

the other hand other offices who really needed new computers did not receive

any” (Respondent 13).

The small budget for ICT investment was only used to support day-to-day

operations (Respondent 4). Respondent 4 further argued that an insignificant budget

was allocated to future systems development. Systems were developed in an ad-hoc

fashion and based on the need to complement what was already currently running.

6.5.5 Resource Complementarity

The Indonesian Treasury focused on the development of SPAN. However,

Respondent 1 was concerned about program sustainability because the Indonesian

Treasury lacked people who could formulate the ICT strategic management.

Moreover, it was evident that the Indonesian Treasury did not have qualified human

resources to make the most of the hardware and software purchased (Respondent 3).

The same respondent further gave the example:

“We bought data warehousing and business intelligence software. Having

bought the software we realised that we did not have qualified staff to make the

most of it”

Notwithstanding the availability of the Human Capital Development Plan

(HDCP) to synergistically support ICT development (Respondent 14) by sending staff

to study further in the area of ICT and law in addition to economics and public policy

(Respondent 11), Respondent 11 argued that:

134

“It was not clear as to how they would be utilised”.

The Indonesian Treasury frequently sent people for ICT training but most of

them were not available because they were not ready or they were transferred to other

offices.

At the operational level, the Indonesian Treasury offices, the TROs and TSOs,

viewed the current resources being relevant to achieving the Indonesian Treasury

objectives. Respondent 6 stated:

“whenever headquarters deployed new systems or hardware, staff from TROs

and TSOs were given training to run them”.

Hence daily operation in the TROs and TSOs were manageable.

6.5.6 E-Government Systems Performance

This study found that the Indonesian Treasury was at the initial stage of changing

ICT to be a partner for business (Respondent 8). ICT had long been viewed as tools

only (Respondent 8; Respondent 10). Nevertheless, at the operational level, the

Indonesian Treasury e-Government systems are considered satisfactory (Respondent

3; Respondent 6; Respondent 9). Respondent 7 mentioned that current systems were

considered reliable in supporting daily operations at the TROs and TSOs. However,

the systems had not been satisfying the people at the strategic level, as stated by

Respondent 1:

“The current Indonesian Treasury e-Government systems do not fully deliver

what is needed by the people at the strategic level. However, the systems are

considered satisfactory at the operational level.”

Another response from Respondent 3 conformed:

“From the operational point of view, current systems are considered

satisfactory. The systems can help generate complex government financial

reports. However, I’m questioning whether or not the data is valid, useful and

135

available. For example, when the data is converted to meet SPAN’s standard,

we found some invalid data.”

The systems were considered satisfactory because the Indonesian Treasury did

not have high expectations as stated by Respondent 1:

“I can see that the user’s interests are mainly in getting the job done. The

expectation is not high.”

Respondent 1 further stated:

“The Indonesian Treasury e-Government performance is better than the manual

one. However, if I look at the spending in ICT investment, the systems should

deliver more. One of the issues that I would like to address is in improving the

system reliability.”

Although the law mandated that public financial management should be run as

an integrated system (Respondent 4), the Indonesian Treasury was aware that current

systems were fragmented, as Respondent 2 mentioned:

“The e-Government performance is not easy to measure as they tend to be

separated. Each business unit has its own systems and maintains its own data.”

Fragmented systems were due to responses to interim needs from business

(Respondent 4; Respondent 10). Hence, high-level officials required extra work to

consolidate data through meetings (Respondent 2).

Although the current systems had hastened the Treasury services, the systems

quality had to be improved (Respondent 6; Respondent 9). The quality of data could

be affected by significant back doors in the existing systems (Respondent 12). Thus,

user awareness needs to be enhanced (Respondent 11).

Based on thematic analysis, it can be inferred that the absence of architectural

document can have some impacts on the e-Government system performance.

136

6.6 Documentary Evidence

6.6.1 Government Regulations

This study was not able to locate Enterprise Architecture documents either in the

Indonesian Treasury or the Indonesian Ministry of Finance. However, this study found

fourteen documents associated with ICT as listed in Appendix D. One of them was

issued by the Indonesian Treasury. It was regulation number KEP10/PB/2008

regarding a standard for hardware and software. Unfortunately, the regulation expired

in 2011. This study was not able to locate any document to replace it.

This study analysed ICT related regulations that were used by the Indonesian

Treasury. Some of these documents referred to internationally known standards such

as Software Development Life Cycle (SDLC) in regulation number

351/KMK.01/2011, ITIL in regulation numbers 414/KMK.01/2011 and 64/KMK.01/

2012, and ISO 20000:2011 in regulation number 83/KMK.01/2013.

6.6.2 Administration Mailing Systems

Using the administration mailing systems archive, the study found that the

number of data requests, report requests and reconciliations from the headquarters is

still growing as seen in Figure 6.7. The Indonesian Treasury maintained its mailing

administration in a web application using a MySQL database. The results were

generated from query statements in Structured Query Language (SQL) as seen in

Appendix E.

137

Figure 6.7. The growth of requests from the Indonesian Treasury Headquarters

The administration mailing systems data were used to understand how beneficial

the current Indonesian Treasury e-Government systems are to the high level officials

in the headquarters. The study used permintaan data (request of data), permintaan

laporan (request of report), and rekonsiliasi (reconciliation) as its criteria in the query.

The frequency of requests can be used to infer whether or not the e-Government

systems provided sufficient information and accessibility for the high level officials.

It could be inferred from the trend of such requests that high-level officials in the

Indonesian Treasury required extra effort in fulfilling their needs for information.

Further detailed results of queries applied to the years 2006 to 2013 can be found in

Appendix F.

6.7 Observation

The study observed how the Indonesian Treasury and its stakeholders interact

with the systems during its peak. The peak of financial transactions normally occurs

at the end of the Indonesian fiscal year. The Indonesian fiscal year follows the calendar

0

1000

2000

3000

4000

5000

6000

7000

2006 2007 2008 2009 2010 2011 2012 2013

Request of Report

Request of Data

Reconciliation

138

year (The Central Intelligence Agency, 2014). It follows that the peak for financial

transactions occurs in November and December.

According to the Indonesian Treasury regulation number 42/2013 regarding

transactions at the end of fiscal year, the last day for payment order submission is on

23 December 2013. During the observation, the study did not find significant queues

on that day and also not for a few days before. This means that the systems did not

experience extensive transactions that could lead to major performance issues. The

situation occurred because the head of TSOs asked its spending units to distribute their

payment orders across several days. The TSO head office also set a limit on the number

of payment orders per day for each spending unit.


Most research techniques used in the qualitative analysis demonstrated similar

findings. However, the observation did not give additional findings since the

Indonesian Treasury managed its transaction load by using human controls. Hence,

treasury transactions were fairly distributed before the peak period was reached. The

results of the interviews with the fifteen respondents are reasonably similar to the

hypotheses of the study wherein the presence or absence of EA is affecting the e-

Government performance. Content analysis of the interviews revealed each group of

respondents had different interests. This is consistent with the EA framework

developed by Zachman (1992) acknowledging various interests for different groups of

an enterprise. Thus, the qualitative findings have partially answered the last two

research questions. This study then used quantitative analysis to complete the answers

of its research questions.

139

QUANTITATIVE FINDINGS

7.1 Overview

This chapter discusses the quantitative analysis used in this study. A variety of

statistical techniques were employed. The stages followed in conducting the

quantitative analysis are reflected in the sections of this chapter and illustrated in

Figure 7.1. It begins by developing measures of constructs in the framework. It follows

by presenting the result of data validity tests. Subsequently, it reveals descriptive

statistics for the respondents from the sample data. More advanced statistical analysis

is discussed in a simple mediation analysis using second-order structural equation

modelling (SEM). The analysis aims to validate the proposed model. It then presents

the outcomes of each of the mediators in the research model. This chapter concludes

with the quantitative analysis results.

7.2 Developing the Measures

In developing the measures, this study makes use of a 6-point Likert scale

(Likert, 1932). The neutral value was excluded in the measure due to the characteristics

of the respondents (Trompenaars & Woolliams, 2011) and to introduce systematic

method adjustment (Raaijmakers, 2000). The measures were developed as observed

variables.

This study developed questions for the observed variables to represent the

dimensions in the model. Each dimension has at least five questions. The e-

Government Performance dimension adopted the questions/statements from Gorla

(2012) and Parasuraman et al. (1988). The lists of questions/statements are listed in

Table 7.1 and Table 7.2.

140

Figure 7.1. Quantitative Analysis Flowchart

Developing the measures

Adopting SERVQUAL s items from previous research

Developing items based on the literature

Validating the measures

Panel of Experts Pre-Test

Subject and Data Collection (Paper-based only)

TSOs (531 sets of Questionnaire sent)

TROs (90 sets of questionnaire sent)

Data Screening

Omitting irresponsive respondents(St.Dev and Count Blank)

Handling Missing Data(Expectation maximization/EM)

Data Validity, Reliability and Bias Tests

Construct Validity(R2)

Reliability Test(Cronbach Alpha)

Common Bias(Harman s single

factor test)

Validating the Model using collected data

Model Fit? Modifying the Model

Mediation Analysis and Hypothesis Testing

Validating Each Dimension in the Model

Validating the Overall Model

Descriptive Statistics of Respondents Data

141

Table 7.1. List of Questions/Statements Developed

EA1 I am involved in the process of designing information systems.

EA2 Frequently requested to suggest ways to optimise the system.

EA3 ICT Standard(s) is/are already available.

EA4 Enterprise attributes (such as business, ICT, people) have fully aligned with the strategic missions.

EA5 Mechanisms are available to propose enterprise enhancement.

EA6 Contribute to some parts of information system development only e.g. system testing, training.

EA7 Data are interchangeable with other government institutions.

EA8 ICT infrastructures always comply with the approved standard(s).

EA9 Enterprise performance indicators affect budget allocation.

OA1 Indonesian Treasury's mission is clear.

OA2 Existing business process has significantly represented the strategic goals.

OA3 The system has fully complied with Indonesian treasury business.

OA4 The changes in organizational structure align with the strategic mission.

OA5 Current system needs significant adjustment.

IA1 Current system can directly generate useful information.

IA2 Data management in the existing systems is simple.

IA3 Data are easy to access.

IA4 High-level officials can customize their own reports.

IA5 High-quality information can be easily generated from the existing system.

RP1 ICT investment relies on existing resources.

RP2 There are no idle resources.

RP3 Investment policies are targeted to fill target performance gaps.

RP4 There are no duplicated resources.

RP5 Information on existing resources is accessible for the system architect and/or high-level officials.

RC1 The intention of ICT investment is solely to support Indonesian treasury's strategic mission.

RC2 Although budget is available, no additional resources are needed because current systems are

already optimised.

RC3 Locally purchased resources always align with headquarters' investment.

RC4 Complementarity resources improve service delivery.

RC5 Unnecessary resources are rarely purchased.

Enterprise Architecture (EA)

Resource Complementarity (RC)

Resource Portfolio Optimisation (RP)

Information Availability (IA)

Organisational Alignment (OA)

142

Table 7.2. List of Questions/Statements adopting SERVQUAL

The study conducted several validity tests to ensure the measures were valid,

reliable and consistent.

7.3 Validity Test of the Measures

Del Greco et al. (1987) argue that there are four validity tests that improve the

consistency and the quality of questionnaires. They are content validity to determine

adequacy, face validity to improve the the look of the questionnaire, criterion validity

to increase the level of questionnaire effectiveness and construct validity to conform

with the theory.

Factor

T1 Up-to-date hardware and software applied immediately.

T2 My office has visually appealing physical facilities.

T3 Neat appearance of IS employees.

T4 Physical facilities align with service provided.

RL1 Information Systems (IS) units always fulfil promises.

RL2 IS units show sufficient interest to solve users problems.

RL3 I am highly dependent on IS units.

RL4 IS units provide timely services.

RL5 IS units record users problems accurately.

RS1 Prompt service to users.

RS2 Willingness to help users.

RS3 IS staff available to respond to user requests.

RS4 Frequent termination of unfinished tasks.

A1 IS units’ staff are instilling confidence.

A2 Users feel secure when using the systems.

A3 Courteous interaction with IS users.

A4 Knowledgeable IS employees.

EM1 Individual attention to users.

EM2 Proper behaviour towards users.

EM3 IS units act in best interest of users

EM4 IS staff understands users’ needs.

EM5 Convenient IS support operating hours.

Empathy

e-Government Performance (e-Gov)

Tangible

Reliability

Responsiveness

Assurance

143

This study employed different resources to comply with all validity tests. A draft

of the questionnaire was sent to Indonesian Treasury staffs who were conducting post

graduate studies in Adelaide. This was done after receiveing suggestions from a panel

of experts’ including research supervisors, a statistical consultant at Flinders

University, and other PhD students from Indonesia and other countries.

7.3.1 Panel of Experts

The research received several recommendations to improve the quality of the

questionnaire. The comments received from the experts are categorsied into validity

criteria as presented in Table 7.3. The study took into account some major

contributions to the questionnaire from the experts, such as:

1. The layout: experts advised a redesign of the initial questionnaire to make it

look professional;

2. The order of questions: experts suggested that the order of questions be

randomised to make the answers less predictable;

3. Breaking questions into groups: it was suggested that questions be collected

into two or three groups per page to guard against participants from becoming

uninterested in responding to the questions/statements;

4. Selection of words: the Indonesian PhD students advocated for the most

commonly accepted interpretation in Indonesian language to minimise

misinterpretation;

The study accomodated the recommendations in the revised questionnaire. The

study then sent them to the Indonesian Treasury employees in Adelaide to pre-test the

questionnaire in order to enhance the quality of questions asked.

144

Table 7.3. Group of validity responses received from experts

Validity

Criteria

Research

Supervisors

Statistical

Consultant

Indonesian

PhD Students

Other PhD

Students

Content √ √ √ √

Face - √ √ -

Criterion √ √ √ √

Construct √ √ - -

7.3.2 Pre-Test

The study received seven responses from ten Indonesian Treasury employees

who were studying in Adelaide. They completed the questionnaires and gave

comments on any area of the questionnaire that could be improved. Their contributions

were significant in improving the questionnaire’s content and criterion validities.

Proper terminology was suggested such as pengadaan (procurement) instead of

investasi (investment). Although those words have different semantic meanings, the

Indonesian Treasury uses procurement interchangebly with investment.

The finalised questionnaire was sent to the Ethics Committee at Flinders

University to gain ethics approval. Having received the ethics approval, the study sent

the questionnaire to the subjects of the research.

7.4 Subject and Data Collection

Since the aim of this study is to investigate ways to improve the e-Government

performance in developing countries with the Indonesian Treasury as its case study,

this study validated its research model with Indonesian Treasury employees. The study

involved the Indonesian Treasury offices nationwide including its headquarters, the

Treasury Regional Offices (TRO), and the Treasury Service Offices (TSO).

The respondents were informed of the expected benefits arising from this study

for the Indonesian Treasury. Consequently, respondents were expected to be careful

and honest in responding to the survey. Although the study did not offer any direct

145

recompense in any form for a response, the number of responses received was around

90% of invitations sent. This indicates that the study was considered important and

useful by the respondents.

Upon receiving responses, the study conducted preliminary data preparation

activities. Data preparation in this study comprises data coding, data entry, and data

cleansing. First, this study prepared the database before entering any data. The number

of fields used in the database was coded according to the observed variables. The

scales in the database referred to the scales in the questionnaire from 1 (strongly

disagree) to 6 (strongly agree).

Second, this study entered the data into statistical package SPSS version 22.

Third, this study conducted a thorough evaluation of the data by looking for maximum

and minimum values for each response. This study manually compared the results with

the original responses. Thus, this study could ensure that the data had been accurately

entered before analysing its descriptive statistics.

7.5 Descriptive Statistics of Respondents Data

According to Pallant (2011), descriptive statistics can be used to show the basic

features of the data, to validate any incorrect asumptions, and to generate further

research questions. Using the frequency feature of SPSS version 22, the study

generated descriptive statistics for each dimension in the research framework and for

the demography of its respondents.

This study received 561 responses. Most responses came from males as seen in

Table 7.4. This response distribution reflects the gender distribution of Indonesian

Treasury employees as shown in Appendix G.

146

Table 7.4. Respondent Gender

Frequency Percent Valid Percent Cumulative Percent

Valid Male 433 77.2 86.9 86.9

Female 65 11.6 13.1 100.0

Total 498 88.8 100.0

Missing System 63 11.2

Total 561 100.0

The majority of responses were received from Indonesian Treasury staff aged

more than 40 years old, as seen in Table 7.5. Respondent’s age distribution is also

deemed similar with employee’s age distribution as presented in Appendix G.

Table 7.5. Respondent Group of Age in years old


Valid Below 25 33 5.9 5.9 5.9

Between 25 and 30 72 12.8 12.9 18.8

Between 30 and 35 56 10.0 10.0 28.9

Between 36 and 40 72 12.8 12.9 41.8

Between 41 and 45 120 21.4 21.5 63.3

Between 46 and 50 107 19.1 19.2 82.4

More than 51 98 17.5 17.6 100.0

Total 558 99.5 100.0

Missing System 3 .5

Total 561 100.0

The majority of the 561 participants have been working for the Indonesian

Treasury for more than ten years as seen in

Table 7.6. This indicates that the participants had sufficient information over the

life of the dynamic e-Government systems there. This also meant that the participants

were aware of any changes after the enactment of public financial laws.

147

Table 7.6. Working Period


Valid Below 5 39 7.0 7.7 7.7

Between 5 and 9 77 13.7 15.2 22.9

Between 10 and 14 87 15.5 17.2 40.1

Between 15 and 19 105 18.7 20.8 60.9

Between 20 and 24 50 8.9 9.9 70.8

More than 25 years 148 26.4 29.2 100.0

Total 506 90.2 100.0


Total 561 100.0

Table 7.7 shows that the majority of participants were from the Treasury

Service Offices. This distribution is also close to the comparison between the total

number of TROs and TSOs as covered in Appendix G

Table 7.7. Working place


Valid Treasury Regional Office

66 11.8 11.9 11.9

Treasury Service Office 490 87.3 88.1 100.0

Total 556 99.1 100.0

Missing System 5 .9

Total 561 100.0

The study received most responses from Indonesian Treasury officers who were

working on the islands of Java, Sumatera and Borneo as seen in Table 7.8. The

concentration of responses in Java and Sumatera reflects the fact that most offices are

situated on these main islands of Indonesia, as stated in Appendix G.

148

Table 7.8. Office Location Area


Valid Java / Bali 175 31.2 33.5 33.5

Sumatera / Borneo 201 35.8 38.4 71.9

Celebes / Papua 83 14.8 15.9 87.8

Other areas 64 11.4 12.2 100.0

Total 523 93.2 100.0


Total 561 100.0

The study aimed for an equal distribution of the 561 participants across their

roles in their office as stated in Section 4.6. The comparison among roles, as depicted

in Table 7.9, is considered to be close to the expected number. Thus, representation of

each role is present in this study.

Table 7.9. Role in Office


Valid Head of Office 81 14.4 19.1 19.1

Division Head 21 3.7 5.0 24.1

Section Head 141 25.1 33.3 57.4

Staff 180 32.1 42.6 100.0

Total 423 75.4 100.0


Total 561 100.0

Table 7.10 indicates that most participants have earned at least a diploma degree.

It can be inferred from this that the participants have sufficient knowledge on the use

of ICT to achive the Indonesian Treasury objectives.

149

Table 7.10. Educational Background


Valid Postgraduate 126 22.5 22.7 22.7

Undergraduate 294 52.4 52.9 75.5

Diploma 115 20.5 20.7 96.2

Others 21 3.7 3.8 100.0

Total 556 99.1 100.0

Missing System 5 .9

Total 561 100.0

7.6 Data Screening

The quality of the data was verified through preliminary data analysis.

Respondents who missed more than 33% of the questions were excluded. Also

excluded were respondents who provided similar answers to all questions, as indicated

by using the standard deviation for each respondent. Respondents with a standard

deviation less than 0.3 were deemed to not be engaged in the research. The study

rejected nine respondents’ data. They were removed as part of cautious data

preparation before handling missing data in responses.

7.6.1 Handling Missing Data

Missing data could harm further analysis for the estimation of the structural

equation model (SEM) used in this study (Allison, 2003; Davey, 2009; Graham, 2012).

This study employed an expectation maximisation (EM) algorithm to resolve missing

data. The EM algorithm is one of many techniques that can be used to account for

missing data (Graham, 2012). The EM algorithm is designed to produce the maximum

likelihood (ML) parameter estimates in each variable.

Missing data occurred at random in the responses. As seen in Table 7.11, the

percentage of missing data varies across the questionnaire statements. Five variables

had no missing responses and one variable (EA9) had considerably higher number of

150

missing values at 7.49%. This study used SPSS to conduct missing value analysis

using the EM algorithm.

Table 7.11. Total Missing Responses for Observed Variables

Missing Response(s)

PCT Missing

Number of Variable(s)

Variable(s)

0 0.00% 5 T2,RL1,RL3,A3,EM3

1 0.18% 1 RL5

2 0.36% 4 RS2,A1,EM1,EM5

3 0.53% 3 RC4,EA2,EA8

4 0.71% 5 T1,RS4,EM2,IA1,RP3

5 0.89% 5 IA2,RP4,RC1,RC3,EA4

7 1.25% 2 EM4,IA5

9 1.60% 2 A2,RP5

10 1.78% 3 T4,RS3,EA5

11 1.96% 3 RL2,RL4,OA3

12 2.14% 1 A4

13 2.32% 1 OA5

14 2.50% 3 OA2,IA4,RC2

15 2.67% 2 OA4,EA1

16 2.85% 2 RP2,RC5

17 3.03% 2 OA1,RP1

19 3.39% 2 T3,IA3

20 3.57% 2 EA3,EA7

25 4.46% 1 RS1

42 7.49% 1 EA9

This study calculated Little's MCAR test: Chi-Square = 4707.088, DF = 4728,

Sig. = 0.583. The study conducted a separate variance t Test for EA9 because it had

more than 5% of values missing. The results of the analysis show all the p values did

not exceed 0.05. Consistent with suggestion from Moss (2009) and Enders (2003), the

result of the missing value analysis using EM can be deemed sufficient.

151

This study replaced missing values by conducting a maximum of 200 iterations

of the EM algorithm in order to improve the likelihood (Neal & Hinton, 1998).

Therefore, the data are considered ready for the preliminary analyses.

7.6.2 Preliminary Data Analyses

This study ran its preliminary data analysis to calculate skewness, kurtosis and

outliers. Kline (2011, p. 54) defines outliers as “scores that are different from the

rest”. Errors in data entry, insufficient instruction or layout of the survey instruments

could be the cause of the outliers (Barnett, 1978; Byrne, 2010; Kline, 2011). The

normality of data could be affected by outliers (Karunasena, 2012). Using Boxplot to

identfy the outliers, as susgested by Hair, Black, Babin, Anderson, and Tatham (2010),

this study found only a few cases with serious outliers. These outliers were excluded

from further analysis.

The kurtosis and skewness for each variable was reviewed as part of the

preliminary analysis. Kurtosis refers to the peakedness or flatness of a data distribution

(Hair et al., 2010; Pallant, 2011). Skewness indicates the symmetry of the data

distribution (Pallant, 2011). The presence of extreme skewness and kurtosis can harm

the SEM analysis (Hair et al., 2010).

This study did not find any extreme skewness or kurtosis in its data. As seen in

Appendix I, the observed variables were not considered to be extremely skewed

because none of them has an absolute value of skewness greater than 3.0 (Kline, 2011).

Similarly, for the kurtosis, none of the variables had a value greater than 7.0 (Byrne,

2010). Hence, the result of the examination of the skewness and kurtosis indicates that

the data for this study are within range for further analysis.

152

7.7 Data Reliability, Validity and Bias Tests

In order to run the SEM analysis smoothly, the model should have good

measurements for each of the latent variables (Hair et al., 2010). A construct validity

test and internal consistency reliability test were conducted to determine if this was the

case.

7.7.1 Construct Validity Test

Constuct validity was tested using convergent validity and discriminant validity.

Items in the same factor are considered to satisfy convergent validity when they

demonstrate high correlation among the measures (Garson, 2013c; Kline, 2011).

Whilst discriminant validity refers to low correlation with other measuring factors

(Alzahrani, 2014; Garson, 2013c; Karunasena, 2012; Kline, 2011; Susanto, 2012).

In assessing the construct validity of the measures, this study used SEM in

AMOS version 22. The latent variables were connected using curve covariance arrows

(Garson, 2013c). This study used multiple regression weight (r2) as the convergent

validity (Garson, 2013c; Hooper et al., 2008). The discriminant validity referred to the

correlation between the items and the factors (Byrne, 2010; Garson, 2013c).

Construct validity was conducted in two stages. First was an assessment of the

validity of the second-order factors in the model. Instead of measuring through direct

observed variables, second-order factors were measured through the related first-order

factors (Byrne, 2010; Kline, 2011). The e-Government Performance was the only

dimension that referred to the second-order factor in the model, as seen in Figure 7.2.

153

Figure 7.2. The Relationship of Latent Variables in the Model

7.7.1.1 The Second-order dimension

The study adopted the measurements from SERVQUAL for the e-Government

Performance (eGov) dimension. This study was aware of SERVQUAL’s unstable

dimensionality (Jiang et al., 2002; Landrum et al., 2009; Myerscough, 2002). As a

second-order factor in the model, e-Government Performance was not measured

directly from the observed variables. Therefore, the e-Government factor was

exogenous. It was measured through details of the covariances in the first-order factors

(Kline, 2011). Thus, should this study require any modification to SERVQUAL, it

should not harm the Confirmatory Factor Analysis of the model in this study.

The construct validity test for the second-order dimension used in the SEM

model is as seen in Figure A. 6 of Appendix J. The study dropped RL3 and RS4 to

EA

IA

RP

RC

OA

eGOV

EM

RL

RS

TA

AS

154

ensure convergent validity. Those variables had squared regression weights (r2) less

than 0.20 as seen in Table 7.12.

Table 7.12. Convergent Validity Analysis for the e-Government Dimension

e-Gov Performance

Obs ← Lat R r2

A1 ← AS 0.731 0.535

A2 ← AS 0.701 0.492

A3 ← AS 0.684 0.468

A4 ← AS 0.581 0.338

EM1 ← EM 0.765 0.585

EM2 ← EM 0.528 0.279

EM3 ← EM 0.776 0.602

EM4 ← EM 0.722 0.522

EM5 ← EM 0.462 0.213

RL1 ← RL 0.733 0.537

RL2 ← RL 0.767 0.589

RL3 ← RL 0.305 0.093

RL4 ← RL 0.727 0.528

RL5 ← RL 0.657 0.432

RS1 ← RS 0.663 0.439

RS2 ← RS 0.510 0.260

RS3 ← RS 0.587 0.345

RS4 ← RS -0.297 0.088

T1 ← TA 0.592 0.351

T2 ← TA 0.629 0.396

T3 ← TA 0.632 0.400

T4 ← TA 0.638 0.407

Although the remaining twenty items were statistically significant (p value

<0.01), the observed variables could not be reduced into five factors as seen in Table

7.13. This indicated inconsistencies between the data and the model. As stated earlier,

155

this was consistent to other studies adopting SERVQUAL in the Information Systems

field (Jiang et al., 2002; Landrum et al., 2009; Myerscough, 2002). Consequently this

study needed to reconstruct the factors for SERVQUAL by using Exploratory Factor

Analysis (EFA).

Table 7.13. Factor Score Weight

Variables TA RS AS EM RL

A1 0.089 0.105 0.045 0.065 0.080

A2 0.076 0.089 0.038 0.055 0.068

A3 0.092 0.108 0.046 0.066 0.083

A4 0.058 0.069 0.029 0.042 0.052

EM1 0.044 0.225 0.083 0.091 0.074

EM2 0.020 0.103 0.038 0.042 0.034

EM3 0.043 0.219 0.081 0.089 0.072

EM4 0.038 0.192 0.071 0.078 0.063

EM5 0.010 0.051 0.019 0.021 0.017

RL1 0.059 0.120 0.066 0.047 0.065

RL2 0.081 0.166 0.090 0.064 0.090

RL4 0.060 0.123 0.067 0.048 0.067

RL5 0.042 0.086 0.047 0.033 0.047

RS1 -0.017 -0.201 0.066 0.109 0.092

RS2 -0.010 -0.121 0.039 0.066 0.055

RS3 -0.014 -0.165 0.054 0.090 0.076

T1 0.072 -0.011 0.037 0.014 0.029

T2 0.109 -0.017 0.055 0.021 0.044

T3 0.120 -0.019 0.060 0.023 0.049

T4 0.113 -0.018 0.057 0.022 0.046

The observed variables in the e-Government Performance dimension were

reconstructed using EFA. DeCoster (1998) argues that performing EFA to resolve

inconsistencies between the data and the model is acceptable. However, EFA should

be performed using not more than half of the data (DeCoster, 1998). As such, forty

156

percent of the data was randomly selected using SPSS. More than 200 data sets as a

result of this selection were sufficient to be analysed using factor analysis (Gorsuch,

1997; Henson & Roberts, 2006; Tabachnick & Fidell, 2007).

The data were then tested in SPSS version 22 and MPLUS for the remaining

twenty observed variables. As a result, another observed variable, namely T4, was

dropped. Both MPLUS and SPSS showed that this observed variable had a factor score

weight similar to more than two other factors. Using maximum likelihood, SPSS

suggested three factors for the nineteen observed variables. MPLUS suggested four

factors as the construct in the model. The comparison between the two can be seen in

Table 7.14 with items for each factor highlighted in bold.

Table 7.14. EFA Results Comparison

Observed Variables

MPLUS - 4 FACTORS SPSS (EFA - ML)

1 2 3 4 1 2 3

T1 0.580 0.2 0.0 -0.1 0.434

T2 0.830 0.0 -0.1 0.0 0.707

T3 0.607 0.0 0.1 0.1 0.543

RL1 0.2 0.680 0.0 0.0 0.630

RL2 0.0 0.2 0.2 0.576 0.547

RL4 0.0 0.535 0.2 0.2 0.469

RS1 0.0 0.862 0.0 0.0 0.811

RS2 0.1 -0.2 0.753 0.1 0.708

RS3 0.0 0.2 0.416 0.2 0.525

RL5 0.358 0.3 0.0 0.2 0.372

A1 0.363 0.475 0.1 0.0 0.482

A2 0.422 0.337 0.1 0.0 0.353

A3 0.0 0.0 0.2 0.684 0.613

A4 0.708 0.0 0.0 0.0 0.610

EM1 0.1 0.2 0.763 -0.1 0.616

EM2 0.1 0.0 -0.1 0.767 0.391

EM3 0.465 0.308 0.1 0.1 0.441

EM4 0.0 0.2 0.667 0.0 0.643

EM5 -0.1 0.0 0.518 0.2 0.558

157

The results from MPLUS were chosen for further analysis in the e-Government

Performance dimension. They were selected due to the grouping of observed variables

having close semantic meaning with the languange used for the questionnaire and

because they were relevant to the cultural reactions from the respondents as suggested

by Trompenaars and Woolliams (2011).

The original names of the observed variabls were kept as the constructs of

SERVQUAL. Thus, this study retained Tangible (TA) as factor number 1, Reliability

(RL) as factor number 2 and Assurance (AS) as factor number 4. Empathy and

Responsiveness (EMRS) were merged as factor number 3. The modified e-

Government Performance dimension can be seen in Figure 7.3. All dimensions were

then combined for the overall construct validity test.

158

Figure 7.3. Modified e-Government Performance Dimension

7.7.1.2 All Dimensions in the Model

Having refined the unstable SERVQUAL constructs used for the e-Government

Performance dimension, this study then covaried all dimensions in the model as

depicted in Figure A. 7 of Appendix J. The multiple regression weights (r2) are

presented in Table 7.15. All observed first-order variables of e-Government

performance satisfied the convergent validity test. Thus, no observed variables were

excluded for the e-Government Performance dimension.

159

Table 7.15. Convergent Validity of the Measures

Obs ← Latents R r2

TA ← eGov* 0.960 0.920

RL ← eGov* 0.900 0.810

AS ← eGov* 0.878 0.770

EMRS ← eGov* 0.879 0.770

EM2 ← AS 0.600 0.360

A3 ← AS 0.772 0.600

RL2 ← AS 0.851 0.720

EM5 ← EMRS 0.457 0.210

EM4 ← EMRS 0.784 0.610

EM1 ← EMRS 0.836 0.700

RS3 ← EMRS 0.686 0.470

RS2 ← EMRS 0.528 0.280

RS1 ← RL 0.799 0.640

RL4 ← RL 0.759 0.580

RL1 ← RL 0.806 0.650

A1 ← RL 0.782 0.610

EM3 ← TA 0.813 0.660

A4 ← TA 0.627 0.390

RL5 ← TA 0.694 0.480

T3 ← TA 0.631 0.400

T2 ← TA 0.603 0.360

T1 ← TA 0.585 0.340

A2 ← TA 0.742 0.550

EA1 ← EA 0.443 0.196

EA2 ← EA 0.411 0.169

EA3 ← EA 0.738 0.544

EA4 ← EA 0.539 0.290

Obs ← Latents R r2

EA5 ← EA 0.573 0.328

EA6 ← EA 0.416 0.173

EA7 ← EA 0.366 0.134

EA8 ← EA 0.486 0.236

EA9 ← EA 0.801 0.641

OA1 ← OA 0.654 0.428

OA2 ← OA 0.769 0.592

OA3 ← OA 0.458 0.210

OA4 ← OA 0.696 0.485

OA5 ← OA 0.032 0.001

IA1 ← IA 0.630 0.397

IA2 ← IA 0.228 0.052

IA3 ← IA 0.476 0.227

IA4 ← IA 0.329 0.108

IA5 ← IA 0.727 0.528

RP1 ← RP 0.568 0.323

RP2 ← RP 0.598 0.358

RP3 ← RP 0.585 0.342

RP4 ← RP 0.634 0.402

RP5 ← RP 0.527 0.278

RC1 ← RC 0.661 0.437

RC2 ← RC 0.257 0.066

RC3 ← RC 0.655 0.429

RC4 ← RC 0.558 0.311

RC5 ← RC 0.580 0.336

* indicates second-order factor

160

The number of variables dropped as a result of the convergent validity test were

varied. RP was the only dimension that did not have any observed variables dropped

for further analysis. The study excluded EA1, EA2, EA6 and EA7 for the EA

dimension; OA5 for the OA dimension; IA2 and IA4 for the IA dimension; RC2 for

the RC dimension. The r2 values for the above mentioned variables were less than 0.20.

Variables with an r2 value less than 0.20 indicate high levels of error (Hooper et al.,

2008; Iacobucci, 2009; Susanto, 2012).

This study then ran several tests to ensure discriminant validity. To satisfy

discriminant validity, Garson (2013b) suggests that the correlation values between

observed variables and unintended latent variables should be less than 0.85. In

addition, Schmitt and Stults (1986) suggest that discriminant validity is achieved when

multicollinearity between factors was not present. The correlation value between two

factors should not be more than 1.00, otherwise it indicates the presence of

multicollinearity (Byrne, 2010).

The result of such discriminant validity tests, as seen in Table 7.16 and Table

7.17, were considered to satisfy both criteria mentioned above. This study did not find

any value above 1.00 for the correlation among factors as seen in Table 7.16. Moreover

there was no observed variable with a value more than 0.85 for the unintended factor.

Thus, the findings indicate that discriminant validity had been achieved.

161

Table 7.16. Correlation Between Factors

Pairs of Factors

Correlation Estimate

EA <--> IA 0.803

EA <--> RC 0.747

EA <--> OA 0.770

EA <--> RP 0.818

EA <--> eGov 0.707

IA <--> RC 0.840

IA <--> OA 0.918

IA <--> RP 0.898

IA <--> eGov 0.812

RC <--> OA 0.942

RC <--> RP 0.949

RC <--> eGov 0.802

OA <--> RP 0.893

OA <--> eGov 0.762

RP <--> eGov 0.852

Table 7.17. Correlation Between Items and Factors

Obs RP OA RC IA EA EMRS AS RL TA

RP1 0.065 0.027 0.057 0.034 0.029 0.007 0.008 0.009 0.012

RP2 0.027 0.011 0.024 0.014 0.012 0.003 0.003 0.004 0.005

RP3 0.092 0.039 0.081 0.049 0.042 0.010 0.011 0.012 0.017

RP4 0.034 0.014 0.030 0.018 0.016 0.004 0.004 0.005 0.006

RP5 0.036 0.015 0.032 0.019 0.017 0.004 0.004 0.005 0.007

OA1 0.023 0.101 0.064 0.058 0.020 0.002 0.002 0.002 0.003

OA2 0.026 0.115 0.073 0.066 0.022 0.002 0.003 0.003 0.004

OA3 0.019 0.084 0.053 0.048 0.016 0.002 0.002 0.002 0.003

OA4 0.023 0.101 0.064 0.058 0.020 0.002 0.002 0.002 0.003

RC1 0.053 0.070 0.103 0.012 0.011 0.005 0.006 0.007 0.009

RC3 0.042 0.056 0.083 0.010 0.009 0.004 0.005 0.005 0.007

RC4 0.044 0.059 0.086 0.010 0.010 0.004 0.005 0.005 0.008

RC5 0.033 0.044 0.065 0.008 0.007 0.003 0.004 0.004 0.006

162

IA1 0.033 0.066 0.013 0.128 0.035 0.007 0.008 0.009 0.012

IA3 0.008 0.017 0.003 0.034 0.009 0.002 0.002 0.002 0.003

IA5 0.045 0.090 0.017 0.176 0.047 0.009 0.011 0.012 0.016

EA3 0.011 0.009 0.005 0.014 0.077 0.001 0.001 0.002 0.002

EA4 0.023 0.018 0.010 0.028 0.158 0.003 0.003 0.003 0.005

EA5 0.014 0.011 0.006 0.017 0.096 0.002 0.002 0.002 0.003

EA8 0.022 0.017 0.009 0.027 0.154 0.002 0.003 0.003 0.004

EA9 0.010 0.008 0.004 0.012 0.068 0.001 0.001 0.001 0.002

RS2 0.004 0.001 0.003 0.004 0.002 0.067 0.008 0.009 0.013

RS3 0.006 0.002 0.005 0.006 0.003 0.104 0.013 0.014 0.019

EM1 0.016 0.005 0.013 0.016 0.008 0.263 0.032 0.036 0.049

EM4 0.012 0.004 0.010 0.012 0.006 0.198 0.024 0.027 0.037

EM5 0.003 0.001 0.002 0.003 0.001 0.046 0.006 0.006 0.009

RL2 0.018 0.006 0.015 0.019 0.009 0.032 0.327 0.041 0.057

A3 0.013 0.005 0.011 0.014 0.007 0.024 0.240 0.030 0.042

EM2 0.007 0.002 0.006 0.007 0.003 0.012 0.121 0.015 0.021

RL1 0.010 0.003 0.008 0.011 0.005 0.018 0.021 0.180 0.032

RL4 0.008 0.003 0.007 0.009 0.004 0.015 0.017 0.149 0.027

RS1 0.010 0.003 0.008 0.010 0.005 0.017 0.020 0.173 0.031

A1 0.010 0.003 0.008 0.010 0.005 0.018 0.020 0.175 0.031

T1 0.005 0.002 0.004 0.006 0.003 0.009 0.011 0.012 0.044

T2 0.007 0.003 0.006 0.008 0.004 0.013 0.015 0.017 0.062

T3 0.009 0.003 0.008 0.010 0.005 0.016 0.019 0.021 0.077

RL5 0.009 0.003 0.008 0.010 0.005 0.017 0.019 0.021 0.078

A2 0.013 0.004 0.011 0.013 0.006 0.022 0.026 0.029 0.105

A4 0.010 0.003 0.008 0.010 0.005 0.017 0.020 0.022 0.080

EM3 0.020 0.007 0.016 0.021 0.010 0.035 0.040 0.044 0.162

7.7.2 Reliability Test

It is common to use the Cronbach α (alpha) to identify data reliability (Achmad,

2012; Alzahrani, 2014; Karunasena, 2012; Susanto, 2012). The Cronbach α value

should be above 0.7 (DeVellis, 2003 cited in Pallant, 2011). However, Pallant (2011)

argues that the Cronbach α values are sensitive to the number of items. Low Cronbach

α values with less than ten items can be reported using the mean inter-item correlation

163

(Pallant, 2011). An acceptable range of values for the mean inter-item is between 0.2

and 0.4 (Briggs and Cheek, 1986 cited in Pallant, 2011)

The results of the reliability data test are seen in Table 7.18. All latent variables

are still within the acceptable range of values. Although the Cronbach α value for the

modified IA is less than 0.7, the mean inter-item value is still within the range of 0.2–

0.4. Hence, the data are considered to be reliable for further analysis such as common

method variance.

Table 7.18. Cronbach Alpha and Mean Inter-Item Correlations

Latent Variables

Alpha Mean inter-item

N Items

TA 0.847 - 7

RL 0.865 - 4

EMRS 0.791 - 5

AS 0.786 - 3

OA 0.730 - 4

IA 0.603 0.370 3

RP 0.707 - 5

RC 0.704 - 4

EA 0.715 - 5

7.7.3 Common Method Variance Bias

This study used Harman’s single factor to test potential bias. Harman’s single

factor, a technique using one-factor for all observed varaibles to test common method

variance, has frequently been used in studies (Podsakoff et al., 2003).

Although other techniques such as the marker variable and the multitrait-

multimethod (MTMM) (Malhotra et al., 2006; Podsakoff et al., 2003; Schaller et al.,

2015; Siemsen et al., 2010) were available as an alternatives to test common method

variance, Malhotra et al. (2006) argue that studies in the IS field do not demonstrate

164

significant problems for the method biases. Thus, the use of Harman’s single factor

remained relevant for this study.

The result of the Harman’s single factor, as seen in Table 7.19, is below forty

percent. Hence, it satisfied the requirement for the test, which should not be more than

fifty percent (Gaskin, 2011; Malhotra et al., 2006; Podsakoff et al., 2003). As such, the

data were considered to be ready for the goodness-of-fit analysis with the model.

Table 7.19. Harman's Single Factor Test

Comp. Initial Eigenvalues Extraction Sums of Squared Loadings

Total % of Variance

Cumulative %

Total % of Variance

Cumulative %

1 14.641 36.603 36.603 14.641 36.603 36.603

2 2.06 5.15 41.753

3 1.673 4.184 45.936

4 1.44 3.601 49.537

5 1.302 3.256 52.793

7.8 The goodness-of-fit

Prior to measuring the goodness-of-fit (GOF) of the model, it was important for

the study to apply appropriate estimation methods in AMOS (Byrne, 2010). Selection

of suitable model estimation methods leads to sound measures (Byrne, 2010; Hair et

al., 2010; Kline, 2011; Mulaik, 2009). This study applied maximum likelihood (ML)

to estimate the research model and its data.

There are several methods available for model estimation such as ML,

generalised least square (GLS), weighted least square (WLS), and scale free least

square (Byrne, 2010; Hair et al., 2010; Kline, 2011; Mulaik, 2009; Tabachnick &

Fidell, 2007). ML estimation method has been widely used in recent studies (Bou-

Llusar et al., 2009; Karunasena, 2012; Moshagen, 2012; Preacher, Zhang, & Zyphur,

165

2011; Savalei, 2012; Stromeyer et al., 2014). Hair et al. (2010) argue that current

versions of ML estimation have reliable techniques to resolve data normality issues.

Hence, ML estimation can generate results with high reliability to further validate the

model and its measurements.

The validity of the measurement model in the study was estimated by using

confirmatory factor analysis (CFA), structural equation modelling (SEM) and AMOS

version 22. Byrne (2010) argues that the use of AMOS in conjunction with CFA and

SEM is useful to evaluate the research model based on the sample data. She further

stated that the AMOS program makes it easy to perform CFA due to its graphical

interface and its ability to generate a variety of output analyses in text such as model

fit, number of estimates and modification indices.

The measurement in the model is considered valid when it has reached

acceptable levels of goodness-of-fit (Hair et al., 2010; Kline, 2011; Mulaik, 2009;

Tabachnick & Fidell, 2007). Acceptable levels of GOF means the theory estimated by

the covariance matrix and the reality estimated by the observed covariance matrix are

similar (Hair et al., 2010).

Although several indices were available, this study referred to Chi Square (χ2),

χ2/df, standardised root mean residual (SRMR), root mean square error of

approximation (RMSEA), comparative fit index (CFI), incremental fit index (IFI) and

Tucker-Lewis index (TLI), as seen Table 7.20, to measure the GOF for collected data

in the model.

The GOF should report indices calculating misspecification in the factor

covariance and the factor loading (Hu & Bentler, 1999). Therefore, in addition to Chi

Square (χ2) and χ2/df, the SRMR should always be available in reporting the GOF of

166

SEM model (Bentler, 2007; Hu & Bentler, 1999; Tabachnick & Fidell, 2007). The

SRMR is characterised by its sensitivity to improper factor covariance specification

(Chen, 2007), whereas RMSEA, CFI, IFI and TLI are sensitive to inappropriate factor

loading specification (Chen, 2007; Hu & Bentler, 1999).

Table 7.20. The goodness-of-fit cut-off values (Barrett, 2007; Byrne, 2010; Hu &

Bentler, 1999; Iacobucci, 2010; Marsh et al., 2004; Susanto, 2012)

Indices of Fit Cut-off Values/ Criteria

Chi Square (χ2) The lower the better

χ2/df < 3 Good Fit

3-5 Acceptable

> 5 Poor Fit

SRMR < 0.08 Close Fit

RMSEA < 0.05 Close fit

0.05 - 0.08 Fair Fit

0.08 - 0.10 Mediocre Fit

> 0.10 Poor Fit

CFI, IFI, TLI > 0.95 Good Fit

> 0.90 Acceptable

RMSEA and CFI are the most common indices used to validate SEM model (Hu

& Bentler, 1999; Iacobucci, 2010; Tabachnick & Fidell, 2007). Recent studies in the

e-Government field demonstrated that IFI and TLI were also useful to validate their

models (Alzahrani, 2014; Karunasena, 2012; Susanto, 2012). Thus, the GOF indices

used in the study were considered sufficient to measure the validity of measurements

in the model. Prior to validating the model and the data for each of the hypotheses, the

study began with validating each of the dimensions in the model.

167

7.9 Generating and Validating the Model Dimensions

According to the GOF indices, as seen in Table 7.21, the study found that two

of six dimensions in the model required modification. Highlighted values in Table 7.21

are considered over or below acceptable cut-off values. Hence, the RP and e-Gov

dimensions should be accordingly altered. In altering the model, the study referred to

the modification indices (Byrne, 2010; Tabachnick & Fidell, 2007).

Table 7.21. The GOF of Research Model Dimensions

Indicator EA OA IA RP RC e-GOV

Origin Mod Origin Mod

Chi Square (χ2)

8.849 1.661 0 73.83 2.841 3.684 482.06 194.879

χ2/df 1.770 0.830 0 14.766 0.710 1.842 3.302 2.030

SRMR 0.023 0.010 0.000 0.064 0.013 0.016 0.040 0.028

RMSEA 0.037 0 0.037 0.157 0 0.039 0.064 0.043

CFI 0.992 1 1 0.867 1 0.996 0.939 0.978

IFI 0.992 1 1 0.868 1 0.996 0.939 0.978

TLI 0.984 1 1 0.734 1 0.987 0.928 0.972

Poor fit in the SEM indicated that the relationship between observed and latent

variables were lacking (Whittaker, 2012). As such, further investigation is required.

Chou and Bentler (2002) argue that model generation in the SEM should be viewed as

a temporary model. Thus, it is common to have to modify the model in the SEM (Chou

& Huh, 2012; Sörbom, 1989; Whittaker, 2012). When the model is considered to lack,

the model should be modified and retested (Chou & Bentler, 2002). The modification

indices in the AMOS output can be used as reference to conduct such modification

(Byrne, 2010; Kline, 2011; Whittaker, 2012).

168

7.9.1 The Unchanged Dimensions

The study did not make any modification to the following dimensions: EA, OA,

IA and RC. The standarised estimates for these dimensions are seen in Figure 7.4 and

Figure 7.5.

Figure 7.4. EA and OA Standarised Estimates

Figure 7.5. IA and RC Standarised Estimates

7.9.2 Resource Portfolio Optimisation

The study found that the RP dimension required modification. Almost all

indicators from the GOF (Table 7.21) confirmed it. Only the SRMR value was still

within the acceptable fit value because there was no covariance among residuals in the

RP dimension.

Figure 7.6. Initial RP Standarised Estimates

169

This study covaried the residuals for the RP2 and RP4 variables. Byrne (2010)

suggested that the study could start modifying the model by referring to the highest

values if the model was covaried. Table 7.22 shows that the covariance between the

residuals of RP2 and RP4 can create significant changes. Before covarying the

residuals, the study confirmed this with the list of questions. Respondents might

consider that these corelated items were similar in intent. The perception of “there are

no idle resources” (RP2) may overlap with “there are no duplicated resources” (RP4).

Thus, the study covaried these residuals.

Table 7.22. RP Modification Indices - Covariances

M.I. Par Change

e4 <--> e5 7.096 -.094

e3 <--> e5 16.909 .090

e3 <--> e4 12.958 -.097

e2 <--> e5 7.018 -.083

e2 <--> e4 45.849 .260

e2 <--> e3 4.929 -.053

e1 <--> e3 5.685 .068

e1 <--> e2 5.636 -.098

The model modification significantly improved the GOF for the RP dimension

as seen in Table 7.21. The standrised estimates for the refined RP dimension can be

seen in Figure 7.7. Having re-run the model, AMOS did not suggest any further

modification to the RP dimension.

170

Figure 7.7. Modified RP Standarised Estimates

7.9.3 e-Government Performance

Although most measurements for the e-Gov dimension demonstrated significant

lack of fit, this study still reviewed the modification indices in the AMOS output to

improve its level of confidence. As seen in Table 7.23, residuals for RS1 (e44), RS2

(e45) and T1 (e47) were required to be covaried with others. However, most of them

were to be covaried with other factors. In CFA, residuals of observed varaibles can

only be covaried when they are within the same latent factor (Gaskin, 2011; Kline,

2011). Thus, the above mentioned variables were excluded for further analysis by

releasing their constraints in the model (Chou & Bentler, 2002).

Referring to the same modification indices in Table 7.23, this study covaried the

residuals for A4 (e43) and T2 (e48). Both of them were in the same factor: Tangible

(TA). Respondents may have viewed these items as similar. Hence, the “perception of

knowledgeable IS employees” (A4) might lead to having “visually appealing physical

facilities” (T2). Similarly, the residuals of RL4 (e33) and A1 (e40) in the Reliability

(RL) factor were covaried because respondents might perceive “IS units provide

timely services” (RL4) similar with the “IS units’ staff are instilling confidence” (A1).

171

Table 7.23. e-Gov Dimension Modification Indices – Covariances

Covariances M.I. Dropped/Covaried Variables

e44<-->e31 14.305 RS1

e44<-->RL 9.571

e44<-->e41 6.335

e44<-->AS 5.553

e43<-->e44 8.445

e49<-->e44 9.714

e45<-->AS 7.497 RS2

e45<-->RL 27.218

e31<-->e45 6.04

e38<-->e45 5.901

e39<-->e45 48.429

e42<-->e45 6.008

e44<-->e45 16.405

e46<-->e45 16.063

e49<-->e45 4.956

e47<-->e31 10.653 T1

e47<-->e39 10.096

e47<-->AS 7.484

e47<-->e46 6.57

e48<-->e47 21.669

e33<-->e40 8.205 RL4 <--> A1

e43<-->e48 11.062 A4 <--> T2

As such, model modification significantly improved the GOF for the e-Gov

dimension as seen in Table 7.21. The standarised estimates for the final e-Gov model

can be seen in Figure 7.8. All dimensions were considered ready to be analysed as a

whole SEM model.

172

Figure 7.8. Final e-Gov Dimension Standarised Estimates

7.10 Generating and Validating the Overall Model

Although all dimensions had already demonstrated signficant fit, some

indicators for the GOF of the overall SEM model, as seen in Table 7.24, indicated

some fitness issues. The highlighted values in Table 7.24 were either below or above

the cut-off GOF values. Hence, further analysis was needed using modification indices

in the AMOS output. The residuals’ covariance effects were reviewed thoroughly for

173

the causal dimension (EA), mediators (OA, IA, RP and RC), and the effect dimension

(e-Gov).

Table 7.24. The GOF for Overall Model

Indices Original Final

χ2 1611.326 1145.004

df 613 478

χ2/df 2.629 2.395

SRMR 0.055 0.041

RMSEA 0.054 0.050

CFI 0.895 0.923

IFI 0.895 0.923

TLI 0.885 0.915

Three more variables were dropped as a result of the GOF analysis. Modification

indices were not the only sources in modifying the model. This study further reviewed

the questions used in the survey. Some questions may have been ambiguous to

respondents (Iacobucci, 2009). As seen in Table 7.25, EA5, IA3 and OA3 indicated

the presence of cross-loading between factors. Respondents may have been confused

with the question asking about “mechanisms are available to propose enterprise

enchancement” (EA 5). Respondents might infer a logical connection between the EA

factor and the Empathy Responsiveness factor since proposing enhancement indicated

respondents’ concern.

Next, the question asking about “data are easy to access” (IA3) could be inferred

as being similar with the question asking about “information on existing resources is

accessible” (RP5). Respondents might also see the question asking about “the system

has fully complied with Indonesian Treasury business” (OA3) covering the overall OA

174

factor. As a result, covariance between residuals of OA3 and the OA factor have a

large modification index.

Using the same source of information, the study covaried the residuals for the

OA4 (e40) and OA2 (e42). Respondents may have identified these two items as

referring to a similar idea. The notion of “existing business process has significantly

represented the strategic goals” (OA2) and “the changes in organizational structure

align with the strategic mission” (OA4) may have commonality.

Table 7.25. Overall Modification Indices – Covariances

Covariances M.I. Dropped / Covaried Variables

e5<-->e6 5.14 EA5

e5<-->z4 29.378

e5<-->e7 13.662

e5<-->e47 13.658

e5<-->z1 12.65

e5<-->e33 11.689

e5<-->e40 10.526

e5<-->z9 9.699

e5<-->e42 5.393

e10<-->e5 10.937

e19<-->e5 7.527

e21<-->e5 4.5

e22<-->e5 45.242

e24<-->e5 4.119

e31<-->e44 35.086 IA3

e31<-->e46 6.906

e31<-->z7 6.491

e31<-->e43 6.026

e31<-->e36 5.189

e31<-->z9 4.543

e6<-->e31 8.4

175

e41<-->z6 22.639 OA3

e41<-->e44 18.398

e41<-->e42 16.687

e41<-->e43 10.428

e41<-->z7 7.757

e41<-->z3 6.119

e41<-->e45 4.369

e20<-->e41 5.437

e6<-->e41 9.934

e27<-->e41 4.755

e33<-->e41 4.855

e38<-->e41 5.735

e40<-->e41 14.117

e40<-->e42 18.693 OA4 <--> OA2

The result of such model modification significantly improved the GOF of the

overall model as seen in Table 7.24. All indicators satisfied the GOF cut-off values as

stated earlier in Table 7.20. The standarised estimates for the final overall model can

be found in Appendix K. All in all, the model was considered ready for more in-depth

analysis for the effect of each mediator.

176

Figure 7.9. Standarised Estimates for the Final Structural Equation Model

177

7.11 Mediator Analysis of the Model

7.11.1 An Overview of Simple Mediation Analysis using SEM

Mediation analysis is a statistical technique to investigate how one factor affects

another assisted by a third factor (Hayes, 2013; Iacobucci, 2008). This means “an

independent variable has an effect on the mediator, which then affects the outcome

variable” (Coffman, 2011, p. 357). Iacobucci (2012) proves that SEM is a poweful

tool to analyse such causal relations in mediation analysis.

A causal relationship in mediation analysis can be modeled as one variable each

for the causal, effect and mediator. It is called the simple mediation model. The simple

mediaton analysis is depicted by two relationships as seen in Figure 7.10 (Hayes, 2013;

Iacobucci, 2008). Part 1 shows a direct relationship from a causal variable (X) to an

effect variable (Y). The loading of this relationship is measured by path c. Part 2 shows

the relationship between X and Y mediated by M. The direct effect of X on Y is

measured through path c’. The indirect relationship between X and Y is via mediator

M. The indirect relationship is measured through path a for the X→M and path b for

the M→Y.

Figure 7.10. Conceptual Diagram of Simple Mediation Model (Hayes, 2013, p. 87;

Iacobucci, 2008, p. 2)

X

M

Y c

X Y c

Part 1

Part 2

178

The simple mediation analysis using SEM is deemed relevant to this study. The

causal relationship between EA (a high-level document) (Lankhorst, 2009) and e-

Government Performance, as depicted in the research framework in Chapter 3, is

affected by four mediators. Simple mediation analysis for each mediator should

exclude the influence of other mediators (Gaskin, 2013; Kenny, 2014; Kenny et al.,

1998). Thus, four simple mediation analyses were used to validate the effect of EA on

e-Government Performance. Hence, the result of each mediator analysis can be used

to answer the hypotheses in this study.

To conduct mediation analysis, the study modified its structural model. The

standard estimates for each model are available in Appendix K. This study further

confirmed that the GOF of the modified model with and without each mediator was

still within the acceptable cut-off values as seen in Table 7.26.

Table 7.26. The GOF for Mediation Analysis

Indices EA-OA-eGov EA-IA-eGov EA-RP-eGOV EA-RC-eGOV No Mediator

χ2 435.855 387.721 550.370 441.146 312.630

df 199 180 242 221 145

χ2/df 2.19 2.154 2.274 1.996 2.156

SRMR 0.036 0.033 0.036 0.033 0.033

RMSEA 0.046 0.045 0.048 0.042 0.045

CFI 0.959 0.963 0.950 0.962 0.967

IFI 0.959 0.963 0.950 0.962 0.967

TLI 0.952 0.957 0.942 0.957 0.961

7.11.2 Results for Mediation Analysis of the Model

Baron and Kenny (1986) introduced procedures for how to validate the presence

of mediators in a causal effect model by using the result of its regression analysis.

Since then, there have been several critiques and suggestions for Baron and Kenny’s

179

mediation analysis procedure (Hayes, 2009; Mathieu et al., 2008; Shrout & Bolger,

2002).

Bootstrapping has proven to be a salient means for valid mediation analysis in

SEM (Hayes, 2013; Iacobucci, Saldanha, & Deng, 2007; Mathieu et al., 2008; Shrout

& Bolger, 2002). The use of bootstrapping can improve the level of confidence in the

estimated coefficients (Hair et al., 2010).

Zhao et al. (2010) combined the mediation analysis procedure from Baron and

Kenny (1986) and bootstrapping. They referred to the three steps for mediation

analysis from Baron and Kenny (1986, p. 1176) as follows:

A variable functions as a mediator when it meets the following conditions: (1)

variations in levels of the independent variable significantly account for

variations in the presumed mediator (i.e., Path a), (2) variations in the mediator

significantly account for variations in the dependent variable (i.e., Path b), and

(3) when Paths a and b are controlled, a previously significant relation between

the independent and dependent variables is no longer significant, with the

strongest demonstration of mediation occurring when Path c is zero.

Prior to validating the model throuh mediation analysis, path c in Figure 7.10

should satisfy the significance test (Baron & Kenny, 1986; Zhao et al., 2010). All paths

in Figure 7.10 were estimated using the following equations (Zhao et al., 2010):

(1) 𝑀 = 𝑖1 + 𝑎𝑋 + 𝑒1

(2) 𝑌 = 𝑖2 + 𝑐′𝑋 + 𝑒2

(3) 𝑌 = 𝑖3 + 𝑐𝑋 + 𝑏𝑀 + 𝑒3

Using Baron and Kenny’s equations, Zhao et.al. (2010, p.200) suggests three types

with mediation and two types without mediation as follows:

1. Complementary mediation: Mediated effect (a x b) and direct effect (c) both

exist and point in the same direction.

180

2. Competitive mediation: Mediated effect (a x b) and direct effect (c) both exist

and point in opposite directions.

3. Indirect-only mediation: Mediated effect (a x b) exists, but no direct effect.

4. Direct-only nonmediation: Direct effect (c) exists, but no indirect effect.

5. No-effect nonmediation: Neither direct effect nor indirect effect exists.

This study used 500 bootstrapping samples, similar to the number of data

samples (Shrout & Bolger, 2002). The first step for mediation analysis begins with

validating the confidence level of the dependent variable (e-Gov) on the independent

variable (EA). Figure 7.11 shows that the standarised estimates for direct effect of EA

on e-Gov is significantly different from zero at 0.769 and p=0.004 two-tailed.

Figure 7.11. Direct Effect of EA on e-Government without mediators

The second step led this study to validate the effect of each of the mediators (OA,

IA, RP and RC) on the causal relationship between EA and e-Gov. The significance

of the relationships between EA → M and M → e-Gov were investigated for each M

in OA, IA, RP, and RC. Table 7.27 indicates a positive effect of EA on each mediator.

Similarly, Table 7.27 shows the significance of the mediator on e-Government

performance.

Table 7.27. The Effect of Each Mediator on e-Government Performance

OA IA RP RC

EA -> Med (path a) .793(.004) .892(.004) .884(.004) .835(.004)

Med -> e-Gov (path b) .239(.024) .732(.004) .834(.004) .502(.004)

181

The next step of Baron and Kenny’s mediation analysis procedure resulted in the

calculation of the signficance of the direct effect of EA on e-Gov with and without

each mediator. This study also investigated the significance of the indirect effect of

EA on e-Gov. As seen in Table 7.28, the indirect effect (a x b) for each of the mediators

was significant. Whilst the direct effect (c’) of EA on e-Gov for the two mediators IA

and RP were not significant, for the other two mediators, OA and RC, the direct effect

was significant.

Referring to Zhao et.al. (2010), this study found that the IA and RP mediators

were statistically proven to be fully mediating the positive effect of EA on e-Gov in

the Indonesian Treasury. The OA and RC mediators were statistically shown to be

partially mediating the positive effect of EA on e-Gov in the Indonesian Treasury.

Table 7.28 suggests that although the hypotheses of this study are all considered

supported, the extent to which they are supported is different. This is discussed in

Section 8.6.

Table 7.28. The Significance of Mediator Effect

Relationship Direct Effect (EA to eGov) Indirect Effect

Interpretation

Without Mediator

With Mediator

EA-OA-eGov .769(.004) .580(.024) .189(.024) Sig., Mediated by OA (Partial)

EA-IA-eGov .769(.004) .119(.646) .653(.004) Sig., Mediated by IA (Full)

EA-RP-eGov .769(.004) .035(.864) .737(.004) Sig., Mediated by RP (Full)

EA-RC-eGov .769(.004) .350(.024) .419(.004) Sig., Mediated by RC (Partial)

182


The objective of this chapter was to conduct the quantitative analysis of this

study. It includes the preparation of the research instrument, data management, and

preliminary data analysis. The study also assessed the validity and reliability of its

data. As a result, some observed variables were excluded from further analysis. The

study refined the SERVQUAL for its e-Government dimension. This was consistent

with other studies adopting SERVQUAL in the Information Systems field.

To ensure the GOF of the overall model, the GOF for each dimension was

measured. Using the modification indices from the AMOS output, the study dropped

some observed variables and correlated residuals to improve the GOF. Similarly, the

study modified the overall model by using these two approaches. The study could not

modify the relationship between dimensions in the model because it referred to the

CFA. Thus, only dropping variables and defining covariance between residuals in the

same factor were permissible (Byrne, 2010; Garson, 2013b; Kline, 2011). The final

model was considered fit based on its propensity indicators.

Mediation analysis was carried out to validate the study’s hypotheses. All

hypotheses made in Chapter 3 were statistically proven to be supported. Furthermore,

using procedures from Zhao et al. (2010), the study identified the type of mediators

contained in the data of this study. Mediators IA and RP were considered to be fully

mediating the causal relationship between EA and e-Government. Mediators OA and

RC were considered as complementary or partially mediating the relationship. To

complete the analysis, all findings in Chapter 6 and Chapter 7 are discussed and mixed

in the next chapter.

183

DISCUSSION

8.1 Overview

This chapter presents a synthesis of the study and its outcomes. The results from

both the qualitative and quantitative analysis are mixed and discussed. The results of

the descriptive analysis is added to enhance and to fill in the gaps in interpreting the

findings. As a final point, this chapter presents the role of mediators in intervening in

the effect of EA on the Indonesian Treasury e-Government system performance.

8.2 EA contributions to e-Government

Similar to Shekkerman’s (2004) study, this study perceived that there is

correlation between EA initiatives and e-Government progression. As presented in

Section 2.4, although four developed countries, namely Australia, the Republic of

Korea, Singapore and the United Kingdom, adopted EA in different years, it can be

perceived that they demonstrated similar trends in their e-Government progress.

Therefore, this study aimed to further understand the effect of EA on e-Government.

8.3 The Effect of EA on e-Government

The study built its research model on the EABM (Tamm et al., 2011). Having

reviewed relevant literature, Chapter 3 elucidates how the study built its research

model. The study revealed that EA, a high-level document influenced by

organisational theory and systems theory (Bernard, 2012), does not have a direct effect

on e-Government performance. The relationship between EA and e-Government

performance is intervened by four enablers: organisational alignment, information

availability, resource portfolio optimisation and resource complementarity.

184

In its model, this study used EA maturity to measure the quality of EA. The study

proposed an alternative maturity model by using the meta synthesis approach as

presented in Section 3.5. The study also altered the EABM to measure e-Government

performance by using SERVQUAL. The findings in quantitative analysis led the study

to propose a novel dimensionality of SERVQUAL as seen in Section 7.9.3. The study

collected its empirical evidence of the model in its case study starting from

investigating the presence of EA in the Indonesian Treasury.

8.4 The Presence of EA in the Indonesian Treasury

A mixed method approach was employed to collect empirical evidence in the

Indonesian Treasury. A number of research techniques were used in its qualitative and

quantitative approaches. The use of two approaches improved the quality of findings

(Creswell, 2009; Venkatesh et al., 2013).

This study infers that there is no artefact demonstrating the presence of EA in

the Indonesian Treasury. As presented in Chapter 6, some interviewees stated:

“To date such documents are not available.” (Respondent 1).

“The direction of existing e-Government systems remains unclear as to how they

should be developed because there is no architecture as its guidance.”

(Respondent 3)

Hence, the Indonesian Treasury relies on its ICT people for the e-Governement

systems development, as Respondent 4 stated:

“We never had any ICT related architecture documents. We do not know what

we are going to develop within the next five years. Even at the technical level,

we do not have proper documentation as to how ICT is developed. Technical

documentation for the e-Government systems are also absent. Thus, e-

Government systems development is controlled by only a few people.”

Consistent with the results of the interviews, this study was not able to find any

document as an evidence of the existence of EA artefacts in the Indonesian Treasury.

185

Notwithstanding the findings mentioned above, this study found that the results

of a descriptive analysis for three of nine observed variables for the EA factor were

different than others. As seen in Appendix I, mode and median for EA1, EA2 and EA7

variables were 5 and 4. This indicates that the Indonesian Treasury have mechanisms

to let users be involved in designing and improving the systems. Similar to responses

to the EA7 variable, meetings of high-level officials, as stated by Respondent 2, could

be used to not only improve the quality of reports, but also to let the Indonesian

Treasury exchange its data with other government institutions.

8.5 The Indonesian Treasury e-Government Performance

The Indonesian Treasury has not achieved the expected performance from its e-

Government systems. Repondent 1 stated:

“The Indonesian Treasury e-Government performance is better than the manual

one. However, if I look at the spending in its ICT investment, the systems should

deliver more. One of the issues that I would like addressed is the improvement

of the systems reliability.”

Respondent 1 further mentioned that the e-Governement systems had not

delivered what the high-level officials expected. As several interviewees stated in

Chapter 6, extra effort was needed in order to present information to people at the

strategic levels.

Additionally, Respondent 12 argued that he found significant backdoors in the

existing e-Government systems. As such, he argued that the quality of data could be

harmed. Respondent 12 further discussed:

“Since there is no architectural documentation, both at the Ministry of Finance

level and the Indonesian Treasury, as an e-Government systems development

framework, we do not know how ICT will be utilised in the future. Thus, to some

extent I think our investment in ICT is not efficient.”

186

The findings were considered consistent with the results of descriptive analysis

for final SERVQUAL congeneric models as seen in Appendix I. This study then

combined all congeneric models for all dimensions into a structural equation model to

answer the hypotheses raised in Section 3.7.

8.6 Answers to Hypotheses

To enhance the rigor of these discussions, the study triangulated the results of

the quantitative analysis with the qualitative findings. Using statistical results in

Chapter 7, this study illustrated a causal effect of EA on e-Government performance

for each hypothesis.

The use of simple mediation analysis for each of the performance enablers is

deemed important as revealed in Chapter 3. Otherwise, the aims of this study, to

investigate the effect of EA on e-Government performance could be obscured.

Measuring the effect of EA on performance enablers and the effect of performance

enablers on e-Government performance is considered to be distracting from the aims

of this study.

Nonetheless, the presence of performance enablers in the model remain

important to understand the degree of influence of each mediator as an intervening

variable (Kenny et al., 1998) by measuring its effect loading (β) and significance (p-

value). Stastitical significance is presented in symbols. The following symbols were

used: * to denote significance at the p<.05 level; ** to denote significance at the p<.01

level; *** to denote significance at the p<.001 level.

187

8.6.1 H1: Organisational Alignment Mediates the Positive Effect of EA on e-Government Performance

This study found that organisational alignment partially mediated the positive

effect of EA on e-Government Performance. Figure 8.1 illustrates the standardized

regression coefficient between all dimensions, i.e., EA and e-Gov, EA and OA, and

OA and e-Gov were statistically significant. The standardized indirect effect was

0.189. As stated earlier in Chapter 7, the study tested the significance of this indirect

effect using bootstrapping procedures of 500 bootstrapped samples, and a 95%

confidence interval was computed by determining the indirect effects at the 2.4th and

97.6th percentiles. Thus, the indirect effect was statistically significant. As such, this

means the quantitative data from the Indonesian Treasury has supported the first

hypothesis in the study.

Figure 8.1. The Effect of OA as Mediator

In this study, organisational aligment reflects the extent to which the Indonesian

Treasury respondents share common understanding of its strategic objectives. This

means the absence of EA has been affecting the Indonesian Treasury e-Government

performance and its organisational alignment. As stated in Chapter 6, the study found

that the Indonesian Treasury adopted a balanced score card (BSC) to help align its

188

organisational structure and its objectives. BSC was also further used to measure

performance through KPIs.

Notwithstanding, the use of BSC to govern its organisation, the effectiveness of

BSC was not measured in this study. Lankhorst (2009) argues that EA could be used

together with BSC. The presence of EA could therefore improve the quality of BSC to

ensure organisational alignment. At the same time, the presence of EA could enchance

e-Government performance in the Indonesian Treasury.

The above findings were consistent with the qualitative findings as stated in

Chapter 6, Respondent 4 argued:

“Alignment was a long way away.”

Similarly, Respondent 10 stated

“Common understanding of the Indonesian Treasury vision and mission can be

seen from the e-Government systems development which is considered to be

partial. Users can only see their roles and responsibilities by overlooking others.

The need for a common framework to bring them to the same viewpoint is

needed”.

Such findings in the qualitative analysis are considered similar to the findings in

the quantitative analysis. As seen in Appendix I, the mean, mode and median for the

OA1, OA2 and OA4 variables is below three. Hence, the Indonesian Treasury has yet

to achieve a common understanding of its strategic goals as Bernard (2012) and van

Steenbergen et al. (2011) suggested.

In addition, another interviewee said:

“up to now we have developed the systems based on partial needs. They were

not combined into one integrated systems” (Respondent 15).

Respondent 14 supported the statements by saying:

189

“When the existing e-Government systems were launched, it was found that they

were not a complete fit with the business process requirements”.

These findings are considered in line with the responses for OA3 and OA5. It

can be perceived that the Business-ICT alignment has not been attained. As Lankhorst

(2004) and Ross et al. (2006) argue, the presence of EA can lead to the Business-ICT

alignment. Having Business-ICT aligned, communication issues between the business

and ICT technical parties can be reduced (Kappelman, 2010; Pereira & Sousa, 2005)

in developing the e-Government systems.

Both approaches suggest complete and sound findings. The results presented in

Table 7.28 suggest that the effect of EA is partially mediated by Organisation

Alignment. This study can therefore demonstrate that the first hypothesis is considered

to be fairly supported wherein the absence of EA has been affecting the

underperforming of e-Government systems in the Indonesian Treasury. The

relationship tends to be somewhat intervened by organisational alignment issues.

8.6.2 H2: Information Availability Mediates the Positive Effect of EA on e-Government Performance

The study found that information availability (IA) fully mediated the positive

effect of EA on e-Government Performance. Figure 8.2 illustrates that the standardized

regression coefficient between EA and IA was statistically significant. The

standardized regression coefficient between IA and e-Gov was also statistically

significant. The standardized indirect effect was 0.653. As stated earlier in Chapter 7,

the study tested the significance of this indirect effect using bootstrapping procedures

of 500 bootstrapped samples, and the 95% confidence interval was computed by

determining the indirect effects at the 0.4th and 99.6th percentiles. Thus, the indirect

effect was statistically significant. However, the standardized regression coefficient

between EA and e-Gov was not statistically significant. This indicates when IA is a

190

mediator, it affected the relationship between EA and e-Gov. Hence, it can be inferred

that IA fully mediates the effect of EA on e-Gov. As such, this indicates the

quantitative data from the Indonesian Treasury are considered to have supported the

second hypothesis in this study.

Figure 8.2. The Effect of IA as Mediator

In this study, information availability is used to measure the extent of high-

quality information generated from the Indonesian Treasury e-Government systems.

This study found that the Indonesian Treasury requires additional effort in generating

high-quality information, as Respondent 11 confirmed:

“Causing extra effort and processes to bring together the data into valuable

information in the form of reports”

However, the results from the quantitative study shows that respondents

perceived the data to be manageable. The presence of an ICT supervisor in all treasury

offices (TSOs, TROs), as stated in Appendix G, could lead to this perception. Staff

who hold such roles were given regular training for any updates to the systems

(Respondent 6).

191

Descriptive analysis on responses for the IA1, IA3 and IA5 variables in the

questionnaire suggested that high-quality information is not easy to accomplish. These

findings were similar to the following interview responses:

“I found some discrepancies when comparing data from the Treasury Service

Offices and the headquarters” (Respondent 12).

“The Indonesian Treasury business units separately managed their own data.

This led to multiple stages of data consolidation.” (Respondent 2).

These findings are consistent with the notion that the presence of EA could improve

the quality of information generated by the systems (Bernard, 2012; Finkelstein, 2006;

Venkatesh et al., 2007) and data management (Finkelstein, 2006; Ross et al., 2006;

Spewak & Hill, 1993).

Such conditions have been causing difficulties for high-level officials to generate

their own reports. However, descriptive analysis on responses for question IA4 showed

that most respondents perceived that high-level officials were able to generate their

own reports. Respondents might refer to the fact that the Indonesian Treasury

headquarters have been investing in software, such as data warehousing and business

intelligence, to assist high-level officials (Respondent 3).

Nevertheless, Respondent 11 stated:

“Not all leaders in the Indonesian Treasury are aware of what ICT can deliver,

such as being able to create their own customised reports directly from the

systems”.

Supporting this, the number of requests for reports, data and reconcilation, as

presented in Figure 6.7, is considered to be consistently growing since 2006. As such,

this study can infer that the second hypothesis is supported that the absence of EA

facilitated by lack of information availability could have been significantly

contributing to the underperforming e-Government systems.

192

8.6.3 H3: Resource Portfolio Optimisation Mediates the Positive Effect of EA on e-Government Performance

This study found that resource portfolio optimisation (RP) fully mediated the

positive effect of EA on e-Government Performance. Figure 8.3 illustrates that the

standardized regression coefficient between EA and RP was statistically significant.

The standardized regression coefficient between RP and e-Gov was also statistically

significant. The standardized indirect effect was 0.737. As stated earlier in Chapter 7,

the study tested the significance of this indirect effect using bootstrapping procedures

of 500 bootstrapped samples, and the 95% confidence interval was computed by

determining the indirect effects at the 0.4th and 99.6th percentiles. Thus, the indirect

effect was statistically significant. However, the standardized regression coefficient

between EA and e-Gov was not statistically significant. This indicates that when RP

is used as mediator, it affects the relationship between EA and e-Gov. Hence, it can be

inferred that RP fully mediates the effect of EA on e-Gov. As such, this means the

quantitative data from the Indonesian Treasury has supported the third hypothesis in

the study.

Figure 8.3. The Effect of RP as Mediator

In this study, resource portfolio optimisation refers to the extent to which the

Indonesian Treasury makes the most of its existing resources, invests in relevant

193

resources to close performance gaps, and minimises duplicated resources. Descriptive

analysis on responses to the variables used in the quantitative approach indicated that

respondents perceived that the Indonesian Treasury still faced duplication of resources

(Pereira & Sousa, 2005; Zachman, 1997).

These findings are considered consistent with the results of the qualitative

findings. The following responses as presented in Chapter 6 suggest similar

circumstances:

“When I was working in the TRO, I found that on the one hand some TSOs

received new computers when there was no need for additional computers, and

on the other hand other offices who were really in need of new computers did

not receive any” (Respondent 13).

This study also found that Respondent 12 argued that there was no clear direction

as to how and where to invest in ICT because ICT was still viewed as a tool. This was

consistent with the responses on RP3 in the questionnaire. This means, the Indonesian

Treasury still has difficulties in reducing the cost of unnecessary ICT investment

(Perks & Beveridge, 2002; Ross et al., 2006; Spewak & Hill, 1993). All in all, highly

mediated by resource portfolio optimisation, a difficulty from the absence of EA has

an effect on its e-Government performance. That said, this study determines that the

third hypothesis is supported.

8.6.4 H4: Resource Complementarity Mediates the Positive Effect of EA on e-Government Performance

This study found that resource complementarity (RC) partially mediated the

positive effect of EA on e-Government Performance. Figure 8.4 illustrates that the

standardized regression coefficient between all dimensions, i.e., EA and e-Gov, EA

and RC, and RC and e-Gov were statistically significant. The standardized indirect

effect was 0.419. As stated earlier in Chapter 7, the study tested the significance of this

194

indirect effect using bootstrapping procedures of 500 bootstrapped samples, and the

95% confidence interval was computed by determining the indirect effects at the 0.4th

and 99.6th percentiles. Thus, the indirect effect was statistically significant. As such,

this means the quantitative data from the Indonesian Treasury has supported the fourth

hypothesis in this study.

Figure 8.4. The Effect of RC as Mediator

In this study, resource complementarity refers to the extent to which the

Indonesian Treasury’s resources synergistically support its strategic goals. According

to descriptive analysis on responses to the questions for the RC factor, as presented in

Appendix I, only responses to the RC2 variable were considered different to the other

variables. Respondents perceived that the need for ICT resources was still high. This

finding was consistent with the fact that the number of obselete resources in the

Indonesian Treasury, as seen in Figure 5.7, is considered to be high.

Using responses to RC1, RC4 and RC5 the Indonesian Treasury seem to be

struggling in integrating its resources (Bernard, 2012; Spewak & Hill, 1993; Zachman,

1997). The findings were consistent with the responses from some interviewees, such

as:

“We might not need to have high-end specification computers to run the

business” (Respondent 5).

195

“There were no control as to what computer specifications should be purchased

and how the PCs should be used. We were inclined to purchase the most

sophisticated computers without knowing the real needs” (Respondent 7).

“We bought data warehousing and, business intelligence software. Having

obtained them we realised that we did not have qualified staff to make the most

of it” (Respondent 3).

The absence of qualified resources to run data warehousing and business

intelligence software could harm interoperabilty. Ross et al. (2006) argued that the

presence of EA could promote interoperability to achieve systems integration. Hence,

the problems in resources complementarity could slightly have intervened the negative

effect of the absence of EA on e-Government performance.


This chapter triangulated the findings of the qualitative approach and the results

of quantitative analysis. Each hypothesis was discussed by using the statistical figures,

responses from the interviews and document analysis. Both approaches indicate that

the absence of EA has been causing e-Government underperformance. Furthermore,

the effect was mediated by each of the mediators, i.e., OA, IA, RP and RC. Hence, it

can be considered that the use of mixed methods to obtain a complete picture of the

Indonesian Treasury e-Government systems in this study was attained. The following

chapter discusses the contributions, limitations, policy implication and future research

arising from this study.

196

CONCLUSIONS AND FURTHER RESEARCH

9.1 Overview

This chapter concludes the study by presenting the key contributions made. It

begins by outlining the research aims, summarising the results, and presenting the

academic contributions. It then probes implications and limitations of the study.

Finally, it suggests potential further research arising from this study.

9.2 Research Aims and Objectives

This study sought to improve e-Government performance in developing

countries through Enterprise Architecture (EA) by using the Indonesian Treasury as

its case study. It aimed to achieve the following objectives:

1. Understanding how the Indonesian Treasury e-Government systems

were developed.

2. Understanding the effect of design artefacts such as EA on e-Government

performance.

3. Proposing a model to define the causal relationship between EA and e-

Government performance. Investigating the role of mediators in such a

relationship.

4. Empirically validating the proposed model using simultaneous mixed

method approach from several employee roles in the Indonesian

Treasury.

5. Offering suggestions and strategies to draw relevant policies in

developing e-Government systems.

197

9.3 Summary of the results

As stated in Chapter 1, the study investigated e-Government performance in

developing countries. Developing countries were selected because most of them were

not able to demonstrate sustainable e-Government systems development. The United

Nations reported that the maturity of e-Government for developing countries has

remained identical for more than a decade. A review of the literature for this study

found that improving e-Government performance entails more than just technological

solutions. A holistic view of e-Government is needed. This can be achieved through

EA.

Similar to Shekkerman’s (2004) study, this study perceived that there was a

strong correlation between EA activities and e-Government system development.

Nevertheless, the study was not able to find any developing countries adopting EA as

part of their e-Government. Moreover, the study discovered a lack of e-Government

studies using EA as its lens. Therefore, this study was motivated to fill the gap in the

e-Government studies in developing countries through an EA perspective.

The Indonesian Treasury was used as its case study. The Treasury was selected

because it is the heart of the public financial institutions and could provide a significant

impact to e-Government development in Indonesia. In order to shift to the next level

in the UN maturity model, it should demonstrate an online financial transaction

capability. To achieve this, the Indonesian Treasury should have reliable e-

Government systems. Furthermore, the Indonesian Treasury e-Government systems

should be ready for seamless interoperability with other government systems.

To understand this complex phenomenon, this study decided to divide the study

into several stages. First, the study conducted in-depth study to understand the

198

relationship between EA and e-Government performance. Building upon the EABM,

this study managed to refine the EABM for its research framework model. The

proposed model has six dimensions, i.e., EA, organisational alignment, information

availability, resource portfolio optimisation, resource complementarity and e-

Government performance. The e-Government performance was measured by using

SERVQUAL.

Second, the study decided to conduct empirical studies in Indonesia, a country

which is considered to have underdeveloped e-Government systems. It then narrowed

down its investigation to the Indonesian Treasury. According to the Indonesian

financial laws, the Treasury is the core of its public finances. The complexity of the

Indonesian Treasury e-Government systems ought to be high. Thus, the selection of

the Indonesian Treasury government systems is deemed sufficient to validate the

framework in the study.

The study used a mixed method to obtain a complete picture of the effect of EA

on e-Government performance in the Indonesian Treasury. Both approaches,

qualitative and quantitative, were performed simultaneously. Purposive sampling was

used in selecting interviewees for the qualitative approach because this study required

valid responses from high-level officials who really knew the Indonesian Treasury e-

Government systems. The selection was based on current and previous roles held by

the respondents. Respondent’s profiles were attained through publicly available

sources such as the Internet. Audio recorded interview responses were then compiled

into text. The study conducted context and thematic analysis of the transcripts using

NVivo 10.

199

This study used paper-based questionnaires to obtain responses from different

groups of respondents for its quantitative approach. Achmad (2012) adopted more than

one media in his research at the Indonesian Treasury. He revealed that paper-based

questionnaires had the highest response rate. Hence, this study decided to use paper-

based only. This study sent three sets of questionnaires, together with a postage paid,

self addressed envelope, to all vertical offices in the Indonesian Treasury. Three groups

of people were sampled in each office. The study had more than a 90 percent response

rate.

Collected data and the proposed model were analysed using Structural Equation

Modelling (SEM) approach using AMOS version 22. The findings indicated that the

structural model and data fulfilled several fit indices used in the study. All dimensions

in the model satisfied construct reliability, convergence and discriminant validity.

Causal effects of the relationships were assessed using simple mediation analysis.

This study found that the effect of EA on e-Government performance was

mediated by four mediators. This means the effect of EA on e-Government

performance was either reduced by the mediator or fully controlled through the

mediator. Hence, no changes could be immediately envisaged.

In light of the above, the study found that there is a problem in the Indonesian

Treasury e-Government system perfomance. Using the model developed in this

research, the study shows that:

1. The absence of EA has been affecting the e-Government performance;

2. Each of the mediators used in the model to some extent mediate the

effect of EA on e-Government Performance;

200

3. The level of mediation effect for each mediator used in the research are

different;

4. SERVQUAL dimensions are considered unstable and required

alteration; and

5. Culture is considered to have contributed to the findings significantly.

These were substantiated by both a qualitative and quantitative analysis of data

collected from the Indonesian treasury.

9.4 Answers to Research Questions

As stated in Section 1.4, the main research question in the study is the extent to

which EA can be used to improve the Indonesian Treasury e-Government system

performance. To answer this question the study developed four questions.

Q1. Internationally, has EA implementation demonstrated benefits to e-Government

performance?

The study perceived that EA implementation has demonstrated positive effects

on e-Government performance as discussed in Section 2.4.

Q2. How can the quality or the process of EA development affect e-Government

performance?

The results of in-depth literature review on EA and e-Government showed that

the EABM was the appropriate model to depict such a relationship. However, this

study suggested an alteration to the EABM by using EA maturity to measure EA and

the SERVQUAL to measure e-Government performance. Four performance enablers

in the EABM, i.e., organisational alignment, information availability, resource

portfolio optimisation and resource complementarity, remained the same. Chapter 3

201

discussed in detail the model development. The model was used as the research

framework.

Q3. What is the current state of the Indonesian Treasury e-Government systems? How

can it be improved?

The Indonesian Treasury e-Government systems is briefly outlined in Chapter 5.

Additionally, the results of qualitative analysis in Chapter 6 exposed a more

comprehensive picture. Most interviewees confirmed that the Indonesian e-

Government Treasury systems has not delivered expected value. The study found that

generating high-quality information required additional effort such as through

meetings and comparing multiple sources of data. Investment in ICT was considered

not to have long term vision.

A smaller number of interviewees considered current systems to be satisfactory.

They argued that valuable information could still be generated within an acceptable

time frame. Another interviewee contended that the Indonesian Treasury e-

Government systems were satisfactory. However, there were low expectations of the

systems’ outcomes. He further claimed that current systems were surely better than

manual processes but that the systems had not been delivering the value they should

have been.

Although internationally accepted standards such as BSC, SDLC, ITIL and ISO

20000:2011 were adopted in the regulations, this study found that the Indonesian

Treasury was struggling to cope with its Business-ICT alignment, unnecessary

investment, duplication of resources, and integration among resources. Hence, the

adoption of internationally accepted standards had not improved the Indonesian e-

Government performance. Such findings were considered similar to the results of

202

quantitative analysis presented in Chapter 7 and Appendix I. All in all, it can be

determined that the Indonesian Treasury e-Government is under performing.

Q4. How can EA be used to improve the quality of the Indonesian Treasury e-

Government systems?

This study found that there is a relationship between EA and e-Government

performance. It found that EA was absent in the e-Government systems development

in the Indonesian Treasury. The results of qualitative analysis in Chapter 6 show that

the Indonesian Treasury did not have any direction what would be developed in the

future. As such, the findings were consistent with the results of quantitative analyses

in Chapter 7. This study found that the factor loading (β) is 0.769 with 0.004

significance level for direct relationships between EA and e-Government without

imposing any mediator.

The use of simple mediation analysis is deemed sufficient to achieve the aims of

this study. The relationship of EA and e-Government is intervened by each of the

mediators. This indicates the effect of EA on e-Government performance should be

gradual. The following mediators: organisational alignment, information availability,

resource portfolio optimisation and resource complementarity could impact the

relationship between EA and e-Government. Mediated by each of the mediators, the

absence of EA could have an effect on the Indonesian Treasury e-Government system

performance.

9.5 Contributions of the Study

The results of this study contributes in the understanding of ways to improve e-

Government performance in developing countries through Enterprise Architecture.

Unlike previous e-Government studies that adopted mainly a technology perspective

203

(Alzahrani, 2014; Sharif et al., 2013; Susanto, 2012) this study adopted an EA

perspective. An EA perspective is important because government requires a holistic

approach in developing its e-Government systems (Saha, 2010a). Unfortunately, this

study found that e-Government studies in developing countries using EA as its focus

remains limited.

This study began by proposing a model to illustrate the relationship between EA

and e-Government. The study integrated well-known theory, SERVQUAL, to the

model. Supported by a literature review and empirical evidence, the study contributed

the following:

1. This study refines the EA Benefit Model (EABM) from Tamm et.al. (2011)

and extends it by using SERVQUAL to measure e-Government

performance in its final research framework.

2. The main contribution of this study provides a better understanding of

improving e-Government performance in developing countries through EA,

particularly in Indonesia. This study validated the model by using the

Indonesian Treasury as its case study. Although, the EABM and

SERVQUAL were developed in the Western culture, the findings confirm

that the model can be applied in an Asian culture.

3. Once the model was defined, the study developed measures to validate five

of six dimensions in the model. The study found the measures satisfied the

normality of data. The normality of data is essential before conducting

stuctural equation modeling (SEM) analysis (Hair et al., 2010).

Consequently, the novel measures in the study contributed to

metodhological progression in the field of e-Government development

through EA. The study modified SERVQUAL congeneric models to

204

measure e-Government performance in the Indonesian Treasury as seen in

Table 9.1.

This is consistent with other studies suggesting the unstable dimensionality

of the SERVQUAL (see Jiang et al., 2002; Landrum et al., 2009;

Myerscough, 2002). In its final model as depicted in Figure 7.3, the study

combined the empathy and responsiveness factors. Other factors were kept

with their original names. Respondents in the Indonesian Treasury might

see that empathy should be demonstrated.

Table 9.1. Final SERVQUAL Congeneric Models

Factor Observed Variables

TANGIBLE (TA) My office has visually appealing physical facilities.

Neat appearance of IS employees.

IS units record users problems accurately.

Users feel secure when using the systems.

Knowledgeable IS employees.

IS units act in best interest of users

RELIABILITY (RL) Information Systems (IS) units always fulfil promises.

IS units provide timely services.

IS units’ staff are instilling confidence.

ASSURANCE (AS) IS units show sufficient interest to solve user’s problems.

Courteous interaction with IS users.

Proper behaviour towards users.

EMPATHY AND RESPONSIVENESS

(EMRS)

IS staff available to respond to user requests.

Individual attention to users.

IS staff understands users’ needs.

4. This study employed simple mediation analysis using SEM. CFA was used

to validate measurement in its SEM. This study also adopted a bootstrapping

procedure to estimate the confidence level of indirect effect in mediation

analysis (Hayes, 2013; Hayes & Preacher, 2010). The use of mediation

205

analysis using SEM in the field of e-Government studies is considered

limited. This study has shown that mediation analysis can be used to enrich

the findings. Hence, it can suggest more useful implications.

9.6 Policy Implications

E-Government is more than just making public services available online. It

involves multiple processes before reaching the final user interface. Since most

information-intensive processes take place in the back office of government agencies

(Dawes, 2008) such as the Indonesian Treasury, the e-Government systems in such

agencies should be reliable and sustainable. The systems could therefore present high-

quality information, interoperable with other agencies. Hence, the Indonesian Treasury

should have a holistic design in its e-Government systems.

The use of a design artefact such as Enterprise Architecture is often neglected in

developing e-Government systems, particularly in developing counties. It is evident

from this study that the absence of EA in the Indonesian Treasury has been causing

significant under performance of e-Government systems. The Indonesian Treasury

should start adopting an EA in its e-Government development strategy.

The study also found that the impact of EA on e-Government performance would

be mediated by four mediators, i.e. organisational alignment, information availability,

resource portfolio optimisation and resource complentarity. Hence the Indonesian

Treasury e-Government performance may not immediately benefit from EA adoption.

Rather the EA adoption could assist e-Government implementation to be more robust,

reliable, interoperable, and resource effective.

Other developing countries and other Indonesian government agencies may find

similarities from this in-depth study of the Indonesian Treasury e-Government

206

systems. Hence, instead of reinventing the wheel, this study could be replicated to

further justify whether or not EA would be a fit to improve e-Government

performance.

9.7 Limitations and Further Research

The study concentrated on Indonesian Treasury e-Government systems. This

study pertains primarily to changes that occurred to the systems since the enactment

of the Indonesian financial laws from 2004 to 2013. Data is gathered from Indonesian

Treasury officials. The scope of this study is limited to the specific roles of Treasury

in managing the public expenditure. Therefore, other Treasury functions such as debt

management and asset management were excluded. These functions are run by units

outside of the Indonesian Treasury.

Interpretation of the results should be take into account several limitations. First,

with a limited time frame, the findings should be seen as an initial in-depth analysis of

e-Government practices in a developing country by using a single case study. Although

data from 561 responses from the Indonesian Treasury that has been neglecting EA in

its e-Government systems development generated thoughtful results, further

investigation from more government agencies could further validate the model and its

measures. Sub-group analysis based on different agencies could also be useful and

enrich comparison between agencies.

Second, as EA maturity and e-Government systems across countries vary,

extended research in other countries could enrich applicability of the findings. This

study contributes to the EA and e-Government performance relationships, which was

formerly immature. Further research by involving more respondents from different

207

cultures and agencies, who may have diverse conceptions, could lead to more

established measures.

A longitudinal study that examines how maturity evolves and different important

factors in each phase is needed to analyse the drivers of maturity on different types of

government organisations e.g. national and regional. Wide ranges of respondents could

also create more profound results. It will be useful to conduct follow-up research after

the Indonesian Treasury has adopted EA for its e-Government systems. Hence,

generalisability of the model could be attained.


This study provides a distinctive theoretical contribution from EA and e-

Government perspectives. Using relevant literature in developing the model to

empirically validate theories, this study found that the relationships between EA and

e-Government system performance are not direct. They are mediated by at least four

factors. This study also reveals that in addition to well-known standards such as BSC,

ITIL, and SDLC that have been adopted to improve the organizational alignment,

information availability, and resource management, the government still requires an

EA as a baseline to improve performance and to ensure sustainability and

interoperability of its e-Government systems.

In addition, this study refines the SERVQUAL to measure e-Government

performance. Therefore, this study provides an important theoretical contribution to

the e-Government literature by offering an alternative means for measuring e-

Government systems. Furthermore, this study suggests the critical role of Enterprise

Architecture in improving the quality of e-Government systems in developing

countries.

208

Given that governments are forced to work collaboratively, the presence of EA

becomes critical. However, government should not view EA as a magic formula that

can instantly lift e-Government system performance to another level. Hence, in

addition to EA adoption, government is still required to assign adequate attention on

several factors such as organisation, information and resources in its policy.

This novel theoretical contribution based on far-reaching investigation in

conjunction with its measures could be used as a potential underpinning for future

research in the area of e-Government and Enterprise Architecture.

209

REFERENCES

Achmad, N. F. (2012). Public Expenditure Management Reform In Indonesia: An Assessment Of The Roles Of The Treasury And Other Institutions. (Doctor of Philosopy), Flinders University, Adelaide, Australia.

Adelman, C. (2000). A parallel universe: Certification in the information technology guild. Change: The Magazine of Higher Learning, 32(3), 20-29.

AGIMO. (2007). Australian Government Architecture Reference Models Version 1.0 AGIMO. (2009). Australian Government Architecture Reference Models Version 2.0 Ahn, M. J. (2012). Whither E-Government? Web 2.0 and the Future of E-Government. Web

2.0 Technologies and Democratic Governance, 169-182. Aichholzer, G. (2004). Scenarios of e-Government in 2010 and implications for strategy

design. Electronic Journal of e-Government, 2(1), 1-10. Alghamdi, I. A., Goodwin, R., & Rampersad, G. (2011). E-government readiness assessment

for government organizations in developing countries. Computer and Information Science, 4(3), 3-17.

Alghamdi, I. A., Goodwin, R., & Rampersad, G. (2014). Organizational E-Government Readiness: An Investigation in Saudi Arabia. International Journal of Business and Management, 9(5), 14-24.

Allison, P. D. (2003). Missing data techniques for structural equation modeling. Journal of abnormal psychology, 112(4), 545-557.

Alwadain, A. S. A. (2014). A Model of Enterprise Architecture Evolution. (Doctor of Philosophy), Queensland University of Technology, Brisbane.

Alzahrani, M. E. (2014). An Acceptance Model for Citizen Adoption of Web-Based E-Government Services in Developing Countries: A Field Survey and a Case Study in Saudi Arabia. (Doctor of Philosophy), Flinders University, Adelaide, South Australia.

Annan, K. (2001). Message of the Secretary General of the United Nations. Retrieved 27 May,2014, from http://portal.unesco.org/ci/en/ev.php-URL_ID=5576&URL_DO= DO_TOPIC&URL_ SECTION=201.html

Arraj, V. (2010). ITIL®: the basics. Buckinghampshire, UK. Aversano, L., Grasso, C., & Tortorella, M. (2012). A Literature Review of Business/IT Alignment

Strategies. Procedia Technology, 5(0), 462-474. doi: 10.1016/j.protcy.2012.09.051 Badan Pemeriksa Keuangan. (2009). Laporan Pemantauan Tidak Lanjut Atas Hasil

Pemeriksaan Laporan Keuangan Pemerintah Pusat Tahun 2004-2008. Retrieved 25 May, 2015, from http://www.bpk.go.id/assets/files/lkpp/2009/lkpp_2009_ 1386153263.pdf

Badan Pemeriksa Keuangan. (2014). Laporan Pemantauan Tidak Lanjut Hasil Pemeriksaan Atas Laporan Keuangan Pemerintah Pusat Tahun 2007-2012. Retrieved 25 May, 2015, from http://www.bpk.go.id/assets/files/lkpp/2013/lkpp_2013_1402974046. pdf

Baker, J. (2012). The Technology–Organization–Environment Framework. In Y. K. Dwivedi, M. R. Wade, & S. L. Schneberger (Eds.), Information Systems Theory (Vol. 28, pp. 231-245): Springer New York.

Banger, D. R. (2008). Maturity Assessment for the Enterprise Architecture (EA) Function. Retrieved 31 October, 2011, from http://www.s-ea-t.com/EA-MaturityAssesment. pdf

Banisar, D. (2006). Freedom of information around the world 2006: A global survey of access to government information laws. Privacy International.

Banker, R. D., Datar, S. M., Kemerer, C. F., & Zweig, D. (1993). Software complexity and maintenance costs. Commun. ACM, 36(11), 81-94. doi: 10.1145/163359.163375

210

Barnett, V. (1978). The Study of Outliers: Purpose and Model. Journal of the Royal Statistical Society. Series C (Applied Statistics), 27(3), 242-250. doi: 10.2307/2347159

Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of personality and social psychology, 51(6), 1173.

Barrett, P. (2007). Structural equation modelling: Adjudging model fit. Personality and Individual Differences, 42(5), 815-824. doi: http://dx.doi.org/10.1016/ j.paid.2006.09.018

Basili, V. R., & Perricone, B. T. (1984). Software Errors and Complexity: An Empirical Investigation. Commun. ACM, 27(1), 42-52. doi: 10.1145/69605.2085

Basu, S. (2004). E‐government and developing countries: an overview. International Review of Law, Computers & Technology, 18(1), 109-132. doi: 10.1080/13600860410001674779

Bayo-Moriones, A., Merino-Díaz-de-Cerio, J., Antonio Escamilla-de-León, S., & Mary Selvam, R. (2011). The impact of ISO 9000 and EFQM on the use of flexible work practices. International Journal of Production Economics, 130(1), 33-42. doi: http:// dx.doi.org/10.1016/j.ijpe.2010.10.012

Bazeley, P., & Jackson, K. (2013). Qualitative data analysis with NVivo. Thousand Oaks, CA: Sage Publications Limited.

Bellamy, C., & Taylor, J. A. (1994). INTRODUCTION: EXPLOITING IT IN PUBLIC ADMINISTRATION - TOWARDS THE INFORMATION POLITY? Public Administration, 72(1), 1-13. doi: 10.1111/j.1467-9299.1994.tb00996.x

Bellman, B., & Rausch, F. (2004). Enterprise architecture for e-Government. Springer, LNCS 3183(EGOV 2004), 48-56.

Benbasat, I., Goldstein, D. K., & Mead, M. (1987). The case research strategy in studies of information systems. MIS Quarterly, 369-386.

Bentler, P. M. (2007). On tests and indices for evaluating structural models. Personality and Individual Differences, 42(5), 825-829. doi: http://dx.doi.org/10.1016/j.paid. 2006.09.024

Bernard, S. A. (2012). An introduction to enterprise architecture (3rd ed.). Bloomington, IN: AuthorHouse.

Bertot, J. C., Jaeger, P. T., & Grimes, J. M. (2010). Using ICTs to Create a Culture of Transparency: E-government and Social Media as Openness and Anti-Corruption Tools for Societies. Government Information Quarterly, 27(3), 264-271. doi: http://dx.doi.org/10.1016/ j.giq.2010.03.001

Bigne, E., Moliner, M. A., & Sanchez, J. (2003). Perceived quality and satisfaction in multiservice organisations: the case of Spanish public services. Journal of Services Marketing, 17(4), 420-442.

Bock, C. (2005). Incentives in e-Government. Journal of E-Government, 1(4), 89-98. doi: 10.1300/J399v01n04_06

Bou-Llusar, J. C., Escrig-Tena, A. B., Roca-Puig, V., & Beltrán-Martín, I. (2009). An empirical assessment of the EFQM Excellence Model: Evaluation as a TQM framework relative to the MBNQA Model. Journal of Operations Management, 27(1), 1-22. doi: http://dx.doi.org/10.1016/ j.jom.2008.04.001

Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77-101. doi: 10.1191/1478088706qp063oa

Brown, A., & Patterson, D. A. (2001). To Err is Human. Paper presented at the Proceedings of the First Workshop on Evaluating and Architecting System dependabilitY (EASY '01).

Burke, B. (2010). IT Score for Enterprise Architecture. Retrieved 28 October, 2011, from http://my.gartner.com/portal/server.pt/gateway/

211

Burton-Jones, A., McLean, E. R., & Monod, E. (2011). On approaches to building theories: Process, variance and systems: Working paper, Sauder School of Business, UBC.

Byrne, B. M. (2010). Structural equation modeling with AMOS: Basic concepts, applications, and programming (Second ed.). New York, USA: Routledge.

Cabinet Office. (2005). Transformational Government: Enabled by Technology. Retrieved 27 May, 2013, from http://www.epractice.eu/files/media/media_201.pdf

Cabinet Office. (2010). Government ICT Strategy. Retrieved 1 July, 2013, from http://webarchive.nationalarchives.gov.uk/20100304104621/http://www.cabinetoffice.gov.uk/media/317444/ict_strategy4.pdf

Carter, L., & Bélanger, F. (2005). The Utilization Of E-Government Services: Citizen Trust, Innovation And Acceptance Factors. Information Systems Journal, 15(1), 5-25. doi: 10.1111/j.1365-2575.2005.00183.x

Cater-Steel, A., & Tan, W.-G. (2005). Implementation of IT Infrastructure Library (ITIL) in Australia: Progress and success factors. Paper presented at the 2005 IT Governance International Conference.

Cavaye, A. L. M. (1996). Case study research: a multi-faceted research approach for IS. Information Systems Journal, 6(3), 227-242. doi: 10.1111/j.1365-2575. 1996.tb00015.x

Chan, Y. E., & Reich, B. H. (2007). IT alignment: what have we learned? Journal of Information Technology, 22(4), 297-315.

Chen, D., & Daclin, N. (2006). Framework for enterprise interoperability. Interoperability for Enterprise Software and Applications, 77-88.

Chen, D., & Daclin, N. (2007). Barriers driven methodology for enterprise interoperability. Establishing the foundation of collaborative networks, 453-460.

Chen, F. F. (2007). Sensitivity of Goodness of Fit Indexes to Lack of Measurement Invariance. Structural Equation Modeling: A Multidisciplinary Journal, 14(3), 464-504. doi: 10.1080/ 10705510701301834

Chief Information Officer. (2005). Enterprise Architecture for UK Government. Retrieved 1 July, 2013, from http://tna.europarchive.org/20080727001118/http:/www.cio. gov.uk/documents/cto/ pdf/enterprise_architecture_uk.pdf

Cho, Y. H., & Choi, B.-D. (2004). E-Government to Combat Corruption: The Case of Seoul Metropolitan Government. International Journal of Public Administration, 27(10), 719-735. doi: 10.1081/PAD-200029114

Chorafas, D. N. (2001). Enterprise architecture and new generation information systems. Florida: CRC Press LLC.

Chou, C.-P., & Bentler, P. M. (2002). Model modification in structural equation modeling by imposing constraints. Computational Statistics & Data Analysis, 41(2), 271-287. doi: http://dx.doi.org/ 10.1016/S0167-9473(02)00097-X

Chou, C., & Huh, J. (2012). Model modification in structural equation modeling. Handbook of structural equation modeling, 232-246.

Chua, J. (2012). The e-Transformation Journey of Singapore. In N. K. Hanna & P. T. Knight (Eds.), National Strategies to Harness Information Technology (pp. 41-76): Springer New York.

Ciborra, C. (2005). Interpreting e-government and development: Efficiency, transparency or governance at a distance? Information Technology & People, 18(3), 260-279.

CIO Council. (1999). Federal Enterprise Architecture Framework. Retrieved 5 July 20011, from http://www.cio.gov/Documents/fedarch1.pdf

Clark, V. L. P., & Creswell, J. W. (2011). Designing and conducting mixed methods research: Thousand Oaks, CA.: Sage.

212

Coffman, D. L. (2011). Estimating Causal Effects in Mediation Analysis Using Propensity Scores. Structural Equation Modeling: A Multidisciplinary Journal, 18(3), 357-369. doi: 10.1080/10705511.2011.582001

Cook, M. E. (2000). What Citizens Want from e-Government. Albany, NY: Center for Technology in Government.

Coursey, D., & Norris, D. F. (2008). Models of E-Government: Are They Correct? An Empirical Assessment. Public Administration Review, 68(3), 523-536. doi: 10.1111/j.1540-6210. 2008.00888.x

Creswell, J. W. (2009). Research design: Qualitative, quantitative, and mixed methods approaches (3 ed.): SAGE Publications, Incorporated.

Croasmun, J. T. (2011). Using Likert-Type Scales in the Social Sciences. Journal of Adult Education, 40(1), 19.

Cullen, A. (2012). How to Develop a High-Performance Enterprise Architecture. Retrieved 15 October 2012, 2012, from http://www.cio.com/article/715513/ How_to_Develop_a_High_ Performance_Enterprise_Architecture

Curtin, G. G., Sommer, M. H., & Vis-Sommer, V. (2003). The World of E-Government. Journal of Political Marketing, 2(3-4), 1-16.

Daclin, N., Chen, D., & Vallespir, B. (2006). Enterprise interoperability measurement-Basic concepts. Enterprise Modeling and Ontologies for Interoperability, Luxemburg.

Dada, D. (2006). The failure of e-government in developing countries: A literature review. The Electronic Journal of Information Systems in Developing Countries, 26(7), 1-10.

Danziger, J. N., & Andersen, K. V. (2002). THE IMPACTS OF INFORMATION TECHNOLOGY ON PUBLIC ADMINISTRATION: AN ANALYSIS OF EMPIRICAL RESEARCH FROM THE “GOLDEN AGE” OF TRANSFORMATION [1]. International Journal of Public Administration, 25(5), 591-627.

Davey, A. (2009). Statistical power analysis with missing data: A structural equation modeling approach: Routledge.

Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1989). User acceptance of computer technology: a comparison of two theoretical models. Management Science, 35(8), 982-1003.

Davison, R. M., Wagner, C., & Ma, L. C. K. (2005). From government to e-government: a transition model. Information Technology & People, 18(3), 280-299.

Dawes, S. S. (2008). Introduction to digital government research in public policy and management. In R. Sharda & S. Voß (Eds.), Digital Government (pp. 103-125): Springer.

De Araujo, J. F. F. E. (2001). Improving public service delivery: the crossroads between NPM and traditional bureaucracy. Public Administration, 79(4), 915-932.

DeCoster, J. (1998). Overview of factor analysis. Retrieved 19 August, 2014, from http://www.stat-help.com/factor.pdf

Del Greco, L., Walop, W., & McCarthy, R. H. (1987). Questionnaire development: 2. Validity and reliability. CMAJ: Canadian Medical Association Journal, 136(7), 699.

Dener, C., Watkins, J. A., & Dorotinsky, W. L. (2010). Financial Management Information Systems: 25 Years of World Bank Experience on What Works and What Doesn’t

Depietro, R., Wiarda, E., & Fleischer, M. (1990). The context for change: Organization, technology and environment. The processes of technological innovation, 151-175.

Dix, A., Finlay, J., Abowd, G., & Beale, R. (1997). Human-Computer Interaction: Prentice-Hall, Inc.

Ebrahim, Z., & Irani, Z. (2005). E-government adoption: architecture and barriers. Business Process Management Journal, 11(5), 589 - 611. doi: 10.1108/14637150510619902

Edelenbos, J., & Klijn, E. H. (2006). Managing stakeholder involvement in decision making: a comparative analysis of six interactive processes in the Netherlands. Journal of Public Administration Research and Theory, 16(3), 417.

213

Elhari, K., & Bounabat, B. (2011). Platform for assessing strategic alignment using enterprise architecture: Application to e-government process assessment. IJCSI International Journal of Computer Science, 8(1), 257-264.

Elo, S., & Kyngäs, H. (2008). The qualitative content analysis process. Journal of Advanced Nursing, 62(1), 107-115. doi: 10.1111/j.1365-2648.2007.04569.x

Enders, C. (2003). Using the expectation maximization algorithm to estimate coefficient alpha for scales with item-level missing data. Psychol. Methods, 8(3), 322-337. doi: 10.1037/1082-989X.8.3.322

Finger, M., & Pécoud, G. (2003). From e-Government to e-Governance? Towards a model of e-Governance. Electronic Journal of e-Government, 1(1), 1-10.

Finkelstein, C. (2006). Enterprise Architecture For Integration : Rapid Delivery Methods and Technologies. London: ARTECH HOUSE.

Fountain, J. E. (2001). Building the virtual state: Information technology and institutional change: Brookings Institution Press.

Fountain, J. E., & Osorio-Urzua, C. A. (2001). Public sector: Early stage of a deep transformation. In R. E. Litan & A. M. Rivlin (Eds.), The Economic Payoff From the Internet Revolution: Brookings Institution Press.

Fowler, F. J. (2014). Survey research methods (Fifth ed.). Los Angeles: SAGE. Fuller, W. A. (2009). Sampling Statistics. Hoboken, New Jersey: John Wiley & Sons, Inc. Gable, G. G. (1994). Integrating case study and survey research methods: an example in

information systems. European Journal of Information Systems, 3(2), 112-126. Garson, G. D. (2013a). Case Study Research. Asheboro, NC: Statistical Associates Publishers. Garson, G. D. (2013b). Factor Analysis. Asheboro, NC: Statistical Associates Publishers. Garson, G. D. (2013c). Validity & Reliability. Asheboro, NC: Statistical Associates Publishers. Gaskin, J. (2011, 25 March 2015). Confirmatory Factor Analysis. Retrieved 13 April, 2015,

from http://statwiki.kolobkreations.com/wiki/Confirmatory_Factor_Analysis Gaskin, J. (2013). SEM Series Part 8: Mediation. Retrieved 23 April, 2015, from

https://youtu.be/0artfnxyF_A Gerke, L., & Ridley, G. (2006). Towards an abbreviated COBIT framework for use in an

Australian State Public Sector. Paper presented at the 17th Australasian Conference on Information Systems, Adelaide.

Gil-Garcia, J. R. (2005). Enacting state websites: A mixed method study exploring e-government success in multi-organizational settings. Paper presented at the University at Albany, State University of New.

Gil-Garcia, J. R., & Martinez-Moyano, I. J. (2007). Understanding the evolution of e-government: The influence of systems of rules on public sector dynamics. Government Information Quarterly, 24(2), 266-290. doi: http://dx.doi.org/10.1016/ j.giq.2006.04.005

Gorla, N. (2012). Information Systems Service Quality, Zone Of Tolerance, And User Satisfaction. Journal of Organizational and End User Computing, 24(2), 50-73.

Gorsuch, R. L. (1997). Exploratory Factor Analysis: Its Role in Item Analysis. Journal of Personality Assessment, 68(3), 532-560. doi: 10.1207/s15327752jpa6803_5

Graham, J. W. (2012). Missing data: Analysis and design: Springer. Grant, G., & Chau, D. (2006). Developing a Generic Framework for E-Government. In M. G.

Hunter (Ed.), Advanced Topics in Information Management (Vol. 5, pp. 72-94). Hershey, PA: Idea Group Inc (IGI).

Gray, D. E. (2004). Doing research in the real world (First ed.). Thousand Oaks, CA: Sage Publication Ltd.

Gregor, S., Hart, D., & Martin, N. (2007). Enterprise architectures: enablers of business strategy and IS/IT alignment in government. Information Technology & People, 20(2), 96-120. doi: 10.1108/09593840710758031

214

Griffiths, J. (2003). Balanced Scorecard Use in New Zealand Government Departments and Crown Entities. Australian Journal of Public Administration, 62(4), 70-79. doi: 10.1111/j..2003.00350.x

Grönlund, Å. (2008). Lost In Competition? The State of the Art in E-Government Research. In H. Chen, L. Brandt, V. Gregg, R. Traunmüller, S. Dawes, E. Hovy, A. Macintosh, & C. Larson (Eds.), Digital Government (Vol. 17, pp. 61-83): Springer US.

Guijarro, L. (2007). Interoperability frameworks and enterprise architectures in e-government initiatives in Europe and the United States. Government Information Quarterly, 24(1), 89-101. doi: http://dx.doi.org/10.1016/j.giq.2006.05.003

Guo, Y. (2010, 24-26 Aug. 2010). E-Government: Definition, Goals, Benefits and Risks. Paper presented at the 2010 International Conference on Management and Service Science (MASS), Wuhan.

Gupta, M., & Jana, D. (2003). E-government evaluation: A framework and case study. Government Information Quarterly, 20(4), 365-387.

Guy, R. F., & Norvell, M. (1977). The Neutral Point on a Likert Scale. The Journal of Psychology, 95(2), 199-204. doi: 10.1080/00223980.1977.9915880

Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2010). Multivariate data analysis (Vol. 7): Pearson Prentice Hall Upper Saddle River, NJ.

Hald, F. (2006). Enterprise Architecture in Denmark: Trends in business driven IT within the public sector of Denmark. (Master), Copenhagen Business School, Copenhagen. Retrieved from http:// www.gotzespace.dk/flemminghald.pdf

Hanafiah, M. A., & Goodwin, R. (2011). Government Enterprise Architecture: lessons learned from the United States’ FEAF and Australia’s AGA. International Information Systems Conference (ISICO), 1, 97-102.

Hanafiah, M. A., & Goodwin, R. (2012). An Alternative Enterprise Architecture Maturity Model. IJACP-Proceedings of the International Conference on Business Management & Information Systems, 509-517.

Hanafiah, M. A., & Goodwin, R. (2014). Improving e-Government Performance through Enterprise Architecture. In G. Rampersad & F. Patel (Eds.), Technology Innovation Leadership in Development: A Middle East (West Asia) Perspective (pp. 35-58). New York: Nova Publishers.

Hancock, G. R., & Mueller, R. O. (2013). Structural equation modeling: A second course (2nd ed.). Charlotte, NC: Information Age Publishing, Inc.

Haque, M. S. (1997). Incongruity Between Bureaucracy and Society in Developing Nations: A Critique. Peace & Change, 22(4), 432-462. doi: 10.1111/0149-0508.00061

Harmon, P. (2003). Developing an enterprise architecture. Business Process Trends. Retrieved 15 October, 2012, from http://www.bptrends.com/publicationfiles/ Enterprise%20 Architecture%20Whitepaper-1-23-03.pdf

Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical Mediation Analysis in the New Millennium. Communication Monographs, 76(4), 408-420. doi: 10.1080/03637750903310360

Hayes, A. F. (2013). Introduction to Mediation, Moderation, and Conditional Process Analysis : A Regression-Based Approach Retrieved from http://flinders.eblib.com/patron/ FullRecord.aspx?p=1186800

Hayes, A. F., & Preacher, K. J. (2010). Quantifying and Testing Indirect Effects in Simple Mediation Models When the Constituent Paths Are Nonlinear. Multivariate Behavioral Research, 45(4), 627-660. doi: 10.1080/00273171.2010.498290

Heeks, R. (2002). Information systems and developing countries: Failure, success, and local improvisations. The Information Society, 18(2), 101-112.

215

Heeks, R. (2003). Most e-Government-for-development projects fail. How can risks be reduced? Retrieved 5 August, 2011, from http://unpan1.un.org/intradoc/ groups/public/documents/ cafrad/unpan011226.pdf

Heeks, R. (2008). Success and Failure Rates of eGovernment in Developing/Transitional Countries: Overview. Retrieved 20 May, 2015, from http://www.egov4dev.org/ success/sfrates.shtml

Henson, R. K., & Roberts, J. K. (2006). Use of Exploratory Factor Analysis in Published Research: Common Errors and Some Comment on Improved Practice. Educational and Psychological Measurement, 66(3), 393-416. doi: 10.1177/0013164405282485

Hiller, J. S., & Bélanger, F. (2001). Privacy Strategies for Electronic Government. Retrieved 25 March, 2013, from http://wwwpub.zih.tu-dresden.de/~isih/projekte/keb/ literatur/Privacy_ strategies_for_Electronic_Commerce.pdf

Hillman, A. L. (2004). Corruption and public finance: an IMF perspective. European Journal of Political Economy, 20(4), 1067-1077. doi: http://dx.doi.org/10.1016/ j.ejpoleco.2003.09.004

Hite, R. C. (2003). A Framework for Asseing and Improving Enterprise Architecture Management version 1.1. Retrieved 31 October, 2011, from www.gao.gov/ new.items/d03584g.pdf

Hjort-Madsen, K. (2006, 04-07 Jan. 2006). Enterprise Architecture Implementation and Management: A Case Study on Interoperability. Paper presented at the HICSS '06. Proceedings of the 39th Annual Hawaii International Conference on System Sciences, 2006., Hawaii.

Homburg, V. (2004). E-government and NPM: a perfect marriage? Paper presented at the ICEC '04, Sixth International Conference on Electronic Commerce, Delft, The Netherland.

Homburg, V. (2008). Understanding e-government: Information systems in public administration: Routledge Basingstoke, UK.

Hooper, D., Coughlan, J., & Mullen, M. (2008). Structural equation modelling: guidelines for determining model fit. Electronic Journal of Business Research Methods, 6(1), 53-60.

Hoyle, R. H. (2012). Handbook of structural equation modeling. New York: Guilford Press. Hu, L. t., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis:

Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1-55. doi: 10.1080/10705519909540118

Hughes, O. E. (2003). Public Management and Administration: An Introduction (3rd ed.). New York, USA: Palgrave MacMillan.

Iacobucci, D. (2008). Mediation Analysis (Vol. 156). Thousand Oaks, CA: SAGE Publications. Iacobucci, D. (2009). Everything you always wanted to know about SEM (structural equations

modeling) but were afraid to ask. Journal of Consumer Psychology, 19(4), 673-680. doi: http://dx.doi.org/10.1016/j.jcps.2009.09.002

Iacobucci, D. (2010). Structural equations modeling: Fit Indices, sample size, and advanced topics. Journal of Consumer Psychology, 20(1), 90-98. doi: http://dx.doi.org/10.1016/j.jcps. 2009.09.003

Iacobucci, D. (2012). Mediation analysis and categorical variables: The final frontier. Journal of Consumer Psychology, 22(4), 582-594. doi: http://dx.doi.org/10.1016/ j.jcps.2012.03.006

Iacobucci, D., Saldanha, N., & Deng, X. (2007). A Meditation on Mediation: Evidence That Structural Equations Models Perform Better Than Regressions. Journal of Consumer Psychology, 17(2), 139-153. doi: http://dx.doi.org/10.1016/S1057-7408(07)70020-7

ICT Working Group. (2002). Singapore 2012: The Living Digital Hub–where IT Works, Report by the ICT Working Group. Economic Review Committee.

216

Iden, J., & Langeland, L. (2010). Setting the Stage for a Successful ITIL Adoption: A Delphi Study of IT Experts in the Norwegian Armed Forces. Information Systems Management, 27(2), 103-112. doi: 10.1080/10580531003708378

IFEAD, I. F. E. A. D. (2004). The IFEAD Extended Enterprise Architecture Maturity Model (E2AMM). Retrieved 28 October, 2011, from www.enterprise-architecture.info/ Images/E2AF/ E2AMMv2.PDF

Irani, Z. (2002). Information systems evaluation: navigating through the problem domain. Information & Management, 40(1), 11-24. doi: http://dx.doi.org/10.1016/S0378-7206(01)00128-8

Jaap van den, H., Lida, K., Ad, J. J. C. B., Marc, B., & Monique, E. M. v. D. (2005). An ISO 9001 quality management system in a hospital. International Journal of Health Care Quality Assurance, 18(5), 361-369. doi: 10.1108/09526860510612216

Jacobs, B., & Suckling, S. (2007). Assessing customer focus using the EFQM Excellence Model: a local government case. The TQM Magazine, 19(4), 368-378. doi: doi:10.1108/09544780710756250

Jaeger, P. T. (2004). Beyond Section 508: The spectrum of legal requirements for accessible e-government Web sites in the United States. Journal of Government Information, 30(4), 518-533. doi: http://dx.doi.org/10.1016/j.jgi.2004.09.010

Jaeger, P. T., & Bertot, J. C. (2010). Transparency and technological change: Ensuring equal and sustained public access to government information. Government Information Quarterly, 27(4), 371-376. doi: http://dx.doi.org/10.1016/j.giq.2010.05.003

Janssen, M., & Hjort-Madsen, K. (2007). Analyzing enterprise architecture in national governments: The cases of denmark and the netherlands.

Janssen, M., & Kuk, G. (2006). A complex adaptive system perspective of enterprise architecture in electronic government.

Jiang, J. J., Klein, G., & Carr, C. L. (2002). Measuring Information Systems Service Quality: SERVQUAL from the Other Side. MIS Quarterly, 26(2), 145-166.

Johnson, R. B., & Onwuegbuzie, A. J. (2004). Mixed methods research: A research paradigm whose time has come. Educational researcher, 33(7), 14-26.

Joon, S. H. (2009). E-Government of Korea. Retrieved 24 June, 2013, from http://eng.nia.or.kr/english/bbs/download.asp?fullpathname=%5CData%5Cattach%5C201112221611162611%5CE%2DGovernment+of+Korea%2Epdf&filename=E%2DGovernment+of+Korea%2Epdf

Kaisara, G., & Pather, S. (2011). The e-Government evaluation challenge: A South African Batho Pele-aligned service quality approach. Government Information Quarterly, 28(2), 211-221. doi: http://dx.doi.org/10.1016/j.giq.2010.07.008

Kaplan, A. M., & Haenlein, M. (2010). Users of the world, unite! The challenges and opportunities of Social Media. Business Horizons, 53(1), 59-68. doi: http://dx.doi.org/10.1016/j.bushor. 2009.09.003

Kappelman, L. A. (2010). The SIM Guide To Enterprise Architecture. Boca Raton, FL: CRC Press. Karunasena, K. (2012). An investigation of the public value of e-government in Sri Lanka.

(Doctor of Philosophy), RMIT University, Melbourne, Australia. Kavanaugh, A. L., Fox, E. A., Sheetz, S. D., Yang, S., Li, L. T., Shoemaker, D. J., . . . Xie, L. (2012).

Social media use by government: From the routine to the critical. Government Information Quarterly, 29(4), 480-491. doi: http://dx.doi.org/10.1016/j.giq. 2012.06.002

Keating, M. (1998). Public management reform and economic and social development: OECD. Kellough, J. E., & Nigro, L. G. (2006). Dramatic reform in the public service: At-will

employment and the creation of a new public workforce. Journal of Public Administration Research and Theory, 16(3), 447.

217

Kenny, D. A. (2014). Mediation. Retrieved 22 April, 2015, from http://davidakenny. net/cm/mediate.htm

Kenny, D. A., Kashy, D. A., & Bolger, N. (1998). Data Analysis In Social Psychology. In D. Gilbert, S. Fiske, & G. Lindzey (Eds.), The handbook of social psychology (Vol. 1, pp. 233-265): Boston, MA: McGraw-Hill.

Kettinger, W. J., & Lee, C. C. (1994). Perceived Service Quality and User Satisfaction with the information Services Function. Decision Sciences, 25(5/6), 737-766.

Kim, S., Kim, H. J., & Lee, H. (2009). An institutional analysis of an e-government system for anti-corruption: The case of OPEN. Government Information Quarterly, 26(1), 42-50. doi: http://dx.doi.org/10.1016/j.giq.2008.09.002

Kline, R. B. (2011). Principles and practice of structural equation modeling: Guilford press. Koh, C. E., Prybutok, V. R., & Zhang, X. (2008). Measuring e-government readiness.

Information & Management, 45(8), 540-546. doi: 10.1016/j.im.2008.08.005 Kopits, G., & Craig, J. D. (1998). Transparency in government operations. Washington, DC:

International Monetary Fund. Krippendorff, K. (2004). Content Analysis: An Introduction to its Methodology (2nd ed.).

Thousand Oaks, CA: Sage Publications, Inc. Lai, C. S. K., & Pires, G. (2010). Testing of a Model Evaluating e-Government Portal Acceptance

and Satisfaction. Electronic Journal Information Systems Evaluation Volume, 13(1), 35-46.

Lakeman, R. (2008). Qualitative Data Analysis with NVivo. Journal of Psychiatric and Mental Health Nursing, 15(10), 868-868. doi: 10.1111/j.1365-2850.2008.01257.x

Lallana, E. C. (2004). SMS in business and government in the Philippines. ICT4D Monograph Series, 1.

Landrum, H., Prybutok, V., Zhang, X., & Peak, D. (2009). Measuring IS system service quality with SERVQUAL: Users’ perceptions of relative importance of the five SERVPERF dimensions. Informing Science: the International Journal of an Emerging Transdiscipline, 12(1), 17-35.

Lange, M., Mendling, J., & Recker, J. C. (2012). Realizing benefits from enterprise architecture: a measurement model. Paper presented at the Proceedings of the 20th European Conference on Information Systems (ECIS).

Langenberg, K., & Wegmann, A. (2004). Enterprise architecture: What aspects is current research targeting. Laboratory of Systemic Modeling, Lausanne.

Langer, A. M. (2008). Analysis and Design of Information Systems Third Edition (3rd ed.). London, UK: Springer-Verlag London Limited.

Langer, A. M. (2012). Guide to Software Development. New York, NY: Springer-Verlag London Limited.

Lankhorst, M. (2009). Enterprise Architecture at Work: Modelling, Communication and Analysis (Second ed.). Heidelberg: Springer-Verlag Berlin Heidelberg.

Lankhorst, M. M. (2004). Enterprise architecture modelling—the issue of integration. Advanced Engineering Informatics, 18(4), 205-216. doi: 10.1016/j.aei.2005.01.005

Laudon, K. C., & Laudon, J. P. (2011). Essentials of Management Information Systems (Vol. 9). New Jersey: Prentice Hall.

Layne, K., & Lee, J. (2001). Developing fully functional E-government: A four stage model. Government Information Quarterly, 18(2), 122-136. doi: 10.1016/s0740-624x(01)00066-1

Lean, O. K., Zailani, S., Ramayah, T., & Fernando, Y. (2009). Factors influencing intention to use e-government services among citizens in Malaysia. International Journal of Information Management, 29(6), 458-475. doi: http://dx.doi.org/10.1016/ j.ijinfomgt.2009.03.012

218

Lee, J.-K. (2003). A model for monitoring public sector web site strategy. Internet Research, 13(4), 259-266.

Lee, J. W., Jones, P. S., Mineyama, Y., & Zhang, X. E. (2002). Cultural differences in responses to a likert scale. Research in Nursing & Health, 25(4), 295-306. doi: 10.1002/nur.10041

Lei Chang. (1994). A Psychometric Evaluation of 4-Point and 6-Point Likert-Type Scales in Relation to Reliability and Validity. Applied Psychological Measurement, 18(3), 205-215. doi: 10.1177/014662169401800302

Lenk, K. (2002). Electronic Service Delivery - A driver of public sector modernisation. Information Polity, 7(2,3), 87-96.

Liimatainen, K. (2008). Evaluating benefits of government enterprise architecture. Paper presented at the 31st Information Systems Research Seminar in Scandinavia.

Liimatainen, K., Hoffmann, M., & Heikkilä, J. (2007). Overview of Enterprise Architecture work in 15 countries. Retrieved 19 October, 2012, from http://www.vm.fi/vm/en/ 04_publications_ and_documents/01_publications/04_public_management/20071102Overvi/FEAR_ENGLANTI_kokonaan.pdf

Likert, R. (1932). A technique for the measurement of attitudes. Archives of psychology, 140, 5–53.

Lilian Chan, Y. C. (2004). Performance measurement and adoption of balanced scorecards: A survey of municipal governments in the USA and Canada. International Journal of Public Sector Management, 17(3), 204-221. doi: doi:10.1108/09513550410530144

Löfstedt, U. (2005). E-government-assessment of current research and some proposals for future directions. International journal of public information systems, 1(1), 39-52.

Luftman, J. (2004). Assessing Business-IT Alignment Maturity. In W. V. Grembergen (Ed.), Strategies for information technology governance (pp. 99-128). Hershey, PA: Idea Group Publishing.

Malhotra, N. K., Kim, S. S., & Patil, A. (2006). Common Method Variance in IS Research: A Comparison of Alternative Approaches and a Reanalysis of Past Research. Management Science, 52(12), 1865-1883.

Marche, S., & McNiven, J. D. (2003). E-Government and E-Governance: The Future Isn't What It Used To Be. Canadian Journal of Administrative Sciences / Revue Canadienne des Sciences de l'Administration, 20(1), 74-86. doi: 10.1111/j.1936-4490.2003.tb00306.x

Marsh, H. W., Hau, K.-T., & Wen, Z. (2004). In Search of Golden Rules: Comment on Hypothesis-Testing Approaches to Setting Cutoff Values for Fit Indexes and Dangers in Overgeneralizing Hu and Bentler's (1999) Findings. Structural Equation Modeling: A Multidisciplinary Journal, 11(3), 320-341. doi: 10.1207/s15328007sem1103_2

Martin, N. J. (2005). Business Strategy and Information Systems Alignment: A Study of the Use of Enterprise Architectures in Australian Government. (Doctor of Philosophy), The Australian National University, Canberra.

Mathieu, J. E., DeShon, R. P., & Bergh, D. D. (2008). Mediational Inferences in Organizational Research: Then, Now, and Beyond. Organizational Research Methods. doi: 10.1177/1094428107310089

McDermott, P. (2010). Building open government. Government Information Quarterly, 27(4), 401-413. doi: http://dx.doi.org/10.1016/j.giq.2010.07.002

Microsoft. (2014a). Lead the Path to IT Innovation Today. Retrieved 7 January, 2015, from http://www.microsoft.com/en-au/server-cloud/products/windows-server-2003/

Microsoft. (2014b). Windows XP Support Has Ended. Retrieved 7 January, 2015, from http://windows.microsoft.com/en-AU/windows/end-support-help

Mikaelian, T., Nightingale, D. J., Rhodes, D. H., & Hastings, D. E. (2011). Real Options in Enterprise Architecture: A Holistic Mapping of Mechanisms and Types for

219

Uncertainty Management. IEEE Transactions on Engineering Management, 58(3), 457-470. doi: 10.1109/tem.2010.2093146

Minoli, D. (2008). Enterprise architecture A to Z: frameworks, business process modeling, SOA, and infrastructure technology. Boca Raton, FL: Auerbach Publications an imprint of Taylor & Francis Group, LLC.

Mirchandani, D., Johnson Jr, J., & Joshi, K. (2008). Perspectives of citizens towards e-government in Thailand and Indonesia: A multigroup analysis. Information Systems Frontiers, 10(4), 483-497. doi: 10.1007/s10796-008-9102-7

Mongkolnam, P., Silparcha, U., Waraporn, N., & Vanijja, V. (2009, 1-3 Dec. 2009). A Push for Software Process Improvement in Thailand. Paper presented at the Software Engineering Conference, 2009. APSEC '09. Asia-Pacific.

Moon, M. J. (2002). The Evolution of E-Government among Municipalities: Rhetoric or Reality? Public Administration Review, 62(4), 424-433. doi: 10.1111/0033-3352.00196

Moshagen, M. (2012). The Model Size Effect in SEM: Inflated Goodness-of-Fit Statistics Are Due to the Size of the Covariance Matrix. Structural Equation Modeling: A Multidisciplinary Journal, 19(1), 86-98. doi: 10.1080/10705511.2012.634724

Moss, S. (2009). Expectation maximization to manage missing data. Retrieved 6 April, 2015, from http://www.psych-it.com.au/Psychlopedia/article.asp?id=267

Mulaik, S. A. (2009). Linear causal modeling with structural equations. Boca Raton, FL: CRC Press, an imprint of Taylor & Francis Group.

Myers, M. D., & Tan, F. B. (2003). Beyond models of national culture in information systems research. Advanced topics in global information management, 2, 14-29.

Myers, M. D., & Newman, M. (2007). The qualitative interview in IS research: Examining the craft. Information and Organization, 17(1), 2-26. doi: http://dx.doi.org/10.1016/ j.infoandorg. 2006.11.001

Myerscough, M. A. (2002). Information Systems Quality Assessment: Replicating Kettinger And Lee ‘s USISF/SERVQUAL Combination. Paper presented at the Proceedings of the Eighth Americas Conference on Information Systems (AMCIS).

NASCIO. (2003). NASCIO Enterprise Architecture Maturity Model. Retrieved 23 October, 2012, from http://www.nascio.org/publications/documents/NASCIO-EAMM.pdf

Nasution, M. P. (2003). Public Finance Management Reform. Indonesian Innovation Forum. National Information Society Agency. (2011). Korea e-Government: Government EA

(Enterprise Architecture) & e-Government Standard Framework. Retrieved 28 May, 2013, from http:// www.itu.int/ITU-D/asp/CMS/Events/2011/ict-apps/s9_NIA.pdf

Ndou, V. (2004). E-Government for Developing Countries: Opportunities and Challenges. The Electronic Journal of Information Systems in Developing Countries, 18(1), 1-24.

Neal, R., & Hinton, G. (1998). A View of the Em Algorithm that Justifies Incremental, Sparse, and other Variants. In M. Jordan (Ed.), Learning in Graphical Models (Vol. 89, pp. 355-368): Springer Netherlands.

Negash, S. (2004). Business intelligence. The Communications of the Association for Information Systems, 13(1), 177-195.

Newton, N. (2010). The use of semi-structured interviews in qualitative research: strengths and weaknesses. Exploring Qualitative methods, 1(1), 1-11.

Niven, P. R. (2011). Balanced scorecard: Step-by-step for government and nonprofit agencies. Hoboken, New Jersey: John Wiley & Sons.

Noblit, G. W., & Hare, R. D. (1988). Meta-ethnography: Synthesizing qualitative studies (Vol. 11): Sage Publications, Inc.

Obi, T. (2012). The 2012 Waseda University International e-Government Ranking. Retrieved 19 October, 2012, from http://www.cicc.or.jp/english/news/2012/e-GovRanking2012e.PDF

220

OECD. (2003). The e-Government imperative. Retrieved 3 October, 2012, from http://www. keepeek.com/Digital-Asset-Management/oecd/governance/the-e-government-imperative_9789264101197-en

Office of Management and Budget. (2005). Federal Enterprise Architecture Program : EA Assessment Framework 2. Retrieved 31 October, 2011, from http://www.cio.gov/Documents/ OMB_EA_Assessment_Framework_2.pdf

Op't Land, M., Proper, E., Waage, M., Cloo, J., & Steghuis, C. (2009). Enterprise Architecture: Creating Value by Informed Governance. Heidelberg: Springer-Verlag Berlin Heidelberg.

Østergaard Jensen, A. (2010). Government Enterprise Architecture Adoption. (Master of Science), Copenhagen Business School, Copenhagen. Retrieved from http://studenttheses. cbs.dk/bitstream/handle/10417/1736/anders_oestergaard_jensen.pdf?sequence=1

Pallant, J. (2011). SPSS survival manual (4th ed.). Crows Nest, NSW, Australia: Allen & Unwin. Parasuraman, A., Berry, L. L., & Zeithaml, V. A. (1991). Refinement and Reassessment of the

SERVQUAL Scale. Journal of Retailing, 67(4), 420-450. Parasuraman, A., Zeithaml, V., & Berry, L. (1985). A conceptual model of service quality and

its implications for future research. The Journal of Marketing, 41-50. Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1988). SERVQUAL: A Multiple-Item Scale For

Measuring Consumer Perceptions of Service Quality. Journal of Retailing, 64(1), 12-40.

Pardo, T. A., Nam, T., & Burke, G. B. (2012). E-Government Interoperability: Interaction of Policy, Management, and Technology Dimensions. Social Science Computer Review, 30(1), 7-23. doi: 10.1177/0894439310392184

Parnas, D. L. (1990). Education for computing professionals. Computer, 23(1), 17-22. doi: 10.1109/2.48796

Pereira, C. M., & Sousa, P. (2005). Enterprise Architecture: Business and IT Alignment. Paper presented at the ACM Symposium on Applied Computing, Santa Fe, NM.

Peristeras, V., & Tarabanis, K. (2000). Towards an enterprise architecture for public adminstration using a top-down approach. European Journal of Information Systems, 9(4), 252-260. doi: 10.1057/palgrave.ejis.3000378

Perks, C., & Beveridge, T. (2002). Guide to enterprise IT architecture. New York, NY: Springer-Verlag New York, Inc.

Peters, R. M., Janssen, M., & Engers, T. M. v. (2004). Measuring e-government impact: existing practices and shortcomings. Paper presented at the Proceedings of the 6th international conference on Electronic commerce, Delft, The Netherlands.

Petter, S., DeLone, W., & McLean, E. (2008). Measuring information systems success: models, dimensions, measures, and interrelationships. European Journal of Information Systems, 17(3), 236-263.

Pheng, T., & Boon, G. W. (2007). Enterprise Architecture in the Singapore Government. Handbook of Enterprise Systems Architecture in Practice, 129.

Pitt, L. F., Watson, R. T., & Kavan, C. B. (1995). Service Quality: A Measure of Information Systems Effectiveness. MIS Quarterly, 19(2), 173-187.

Podsakoff, P. M., MacKenzie, S. B., Lee, J.-Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: a critical review of the literature and recommended remedies. Journal of applied psychology, 88(5), 879.

Pollard, C., & Cater-Steel, A. (2009). Justifications, Strategies, and Critical Success Factors in Successful ITIL Implementations in U.S. and Australian Companies: An Exploratory Study. Information Systems Management, 26(2), 164-175. doi: 10.1080/10580530902797540

Poulin, D. (2004). Open Access to Law in Developing Countries. First Monday, 9(12), 1-19.

221

Preacher, K. J., Zhang, Z., & Zyphur, M. J. (2011). Alternative Methods for Assessing Mediation in Multilevel Data: The Advantages of Multilevel SEM. Structural Equation Modeling: A Multidisciplinary Journal, 18(2), 161-182. doi: 10.1080/10705511.2011.557329

Pudjianto, B., Zo, H., Ciganek, A. P., & Rho, J. J. (2011). Determinants of e-government assimilation in Indonesia: An empirical investigation using a TOE framework. Asia Pacific Journal of Information Systems, 21(1), 49-80.

Raaijmakers, Q. A. W. (2000). Adolescents’ Midpoint Responses On Likert-Type Scale Items: Neutral Or Missing Values? International Journal of Public Opinion Research, 12(2), 209-217. doi: 10.1093/ijpor/12.2.209

Rainey, H. G. (2009). Understanding and Managing Public Organizations (Fourth ed.). San Francisco, CA: John Wiley & Sons, Inc.

Rauch, J. E., & Evans, P. B. (2000). Bureaucratic structure and bureaucratic performance in less developed countries. Journal of Public Economics, 75(1), 49-71. doi: http://dx.doi.org/ 10.1016/S0047-2727(99)00044-4

Reddick, C. G., & Norris, D. F. (2013). Social media adoption at the American grass roots: Web 2.0 or 1.5? Government Information Quarterly, 30(4), 498-507. doi: http://dx.doi.org/ 10.1016/j.giq.2013.05.011

Reddick, C. G., & Roy, J. (2012). Business perceptions and satisfaction with e-government: Findings from a Canadian survey. Government Information Quarterly, 1-9. doi: 10.1016/ j.giq.2012.06.009

Reece, B. (2006). E-Government Literature Review. Journal of E-Government, 3(1), 69-110. doi: 10.1300/J399v03n01_05

Reithofer, W., & Naeger, G. (1997). Bottom-up planning approaches in enterprise modelling—the need and the state of the art. Computers in Industry, 33(2), 223-235. doi: 10.1016/s0166-3615(97)00028-6

Relyea, H. C. (2002). E-gov: Introduction and overview. Government Information Quarterly, 19(1), 9-35.

Richard, L. B., & Michael, D. M. (2002). Information systems as a reference discipline. MIS Quarterly, 26(1), 1-14.

Ridley, G., Young, J., & Carroll, P. (2004, 5-8 Jan. 2004). COBIT and its utilization: a framework from the literature. Paper presented at the Proceedings of the 37th Annual Hawaii International Conference on System Sciences, Hawaii.

Rohloff, M. (2008). Framework and Reference for Architecture Design. Paper presented at the Americas Conference on Information Systems (AMCIS), Toronto, ON.

Rokhman, A. (2011). E-Government Adoption in Developing Countries; the Case of Indonesia. Journal of Emerging Trends in Computing and Information Sciences, 2(5), 228-236.

Ross, J. W., Weill, P., & Robertson, D. (2006). Enterprise architecture as strategy: Creating a foundation for business execution. Boston, MA: Harvard Business Press.

Rossel, P., Finger, M., & Misuraca, G. (2006). Mobile e-Government options: between technology-driven and user-centric. The electronic Journal of e-Government, 4(2), 79-86.

Saha, P. (2004). Analyzing The Open Group Architecture Framework from the GERAM Perspective. Retrieved 17 August, 2011, from http://www.opengroup.org/ architecture/wp/saha/ TOGAF_GERAM_Mapping.htm

Saha, P. (2009). Architecting the connected government: practices and innovations in Singapore. Paper presented at the IEGOV2009.

Saha, P. (2009). A Methodology for Government Transformation with Enterprise Architecture. In P. Saha (Ed.), Advances in Government Enterprise Architecture (pp. 1-29). Hershey, PA: Information Science Reference (an imprint of IGI Global).

Saha, P. (2010a). Enterprise architecture as platform for connected government: Advancing the Whole of Government Enterprise Architecture Adoption with Strategic (Systems)

222

Thinking. Retrieved 3 October, 2011, from http://unpan1.un.org/intradoc/ groups/public/documents/ unpan/unpan041801.pdf

Saha, P. (2010b). Enterprise architecture as platform for connected government: Understanding the Impact of Enterprise Architecture on Connected Government - A Qualitative Analysis. Retrieved 3 October, 2011, from http://unpan1.un.org/ intradoc/groups/public/documents/ unpan/unpan039390.pdf

Samaratunge, R., Alam, Q., & Teicher, J. (2008). The new public management reforms in Asia: a comparison of South and Southeast Asian countries. International Review of Administrative Sciences, 74(1), 25-46.

Sandeep, M. S., & Ravishankar, M. N. (2014). The continuity of underperforming ICT projects in the public sector. Information & Management, 51(6), 700-711. doi: http://dx.doi.org/ 10.1016/j.im.2014.06.002

Savalei, V. (2012). The Relationship Between Root Mean Square Error of Approximation and Model Misspecification in Confirmatory Factor Analysis Models. Educational and Psychological Measurement, 72(6), 910-932. doi: 10.1177/0013164412452564

Schaller, T. K., Patil, A., & Malhotra, N. K. (2015). Alternative Techniques for Assessing Common Method Variance: An Analysis of the Theory of Planned Behavior Research. Organizational Research Methods, 18(2), 177-206. doi: 10.1177/1094428114554398

Schick, A. (1998). Why most developing countries should not try New Zealand's reforms. The World Bank Research Observer, 13(1), 123-131.

Schlichting, C., & Mason, J. (2004). Certification training and the academy. Journal of Computing Sciences in Colleges, 20(1), 157-167.

Schmitt, N., & Stults, D. M. (1986). Methodology Review: Analysis of Multitrait-Multimethod Matrices. Applied Psychological Measurement, 10(1), 1-22. doi: 10.1177/ 014662168601000101

Schreiber, R., Crooks, D., & Stern, P. N. (1997). Qualitative meta-analysis. In J. M. Morse (Ed.), Completing a qualitative project: Details and dialogue (pp. 311-326). Thousand Oaks, CA: Sage.

Schultze, U., & Avital, M. (2011). Designing interviews to generate rich data for information systems research. Information and Organization, 21(1), 1-16. doi: http://dx.doi.org/10.1016/ j.infoandorg.2010.11.001

Seppanen, V., Heikkila, J., & Liimatainen, K. (2009, 20-23 July 2009). Key Issues in EA-Implementation: Case Study of Two Finnish Government Agencies. Paper presented at the Commerce and Enterprise Computing, 2009. CEC '09. IEEE Conference on.

Shah, H., & Kourdi, M. E. (2007). Frameworks for Enterprise Architecture. IT Professional, 9(5), 36-41. doi: 10.1109/mitp.2007.86

Shahkooh, K. A., & Abdollahi, A. (2007, March 4-9). A strategy-based model for e-government planning. Paper presented at the Proceedings of the International Multi-Conference on Computing in the Global Information Technology (ICCGI'07), Guadeloupe, French Caribbean.

Shahkooh, K. A., Saghafi, F., & Abdollahi, A. (2008, 7-11 April). A Proposed Model for E-government Maturity. Paper presented at the 3rd International Conference on Information and Communication Technologies: From Theory to Applications, 2008 (ICTTA 2008), Damascus, Syria.

Sharif, M. H. M., Davidson, R., & Troshani, I. (2013). Exploring Social Media Adoption in Australian Local Government Organizations. Paper presented at the International Conference on Information Resources Management (CONF-IRM). http://aisel.aisnet.org/cgi/ viewcontent.cgi?article=1027&context=confirm2013

Shekkerman, J. (2004). Trends in Enterprise Architecture. 1.0. Retrieved 6 March, 2012, from http://www.enterprise-architecture.info/Images/EA%20Survey/EA%20Survey%20 2004%20IFEAD.PDF

223

Shim, D. C., & Eom, T. H. (2008). E-Government and Anti-Corruption: Empirical Analysis of International Data. International Journal of Public Administration, 31(3), 298-316. doi: 10.1080/01900690701590553

Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimental studies: new procedures and recommendations. Psychological methods, 7(4), 422.

Siau, K., & Long, Y. (2005). Synthesizing e-government stage models–a meta-synthesis based on meta-ethnography approach. Industrial Management & Data Systems, 105(4), 443-458.

Siemsen, E., Roth, A., & Oliveira, P. (2010). Common Method Bias in Regression Models With Linear, Quadratic, and Interaction Effects. Organizational Research Methods, 13(3), 456-476. doi: 10.1177/1094428109351241

Silcock, R. (2001). What is E-government. Parliamentary Affairs, 54(1), 88-101. doi: 10.1093/pa/54.1.88

Simeon, R. (1999). Evaluating domestic and international web-site strategies. Internet Research, 9(4), 297-308.

Siskos, E., Askounis, D., & Psarras, J. (2014). Multicriteria decision support for global e-government evaluation. Omega, 46, 51-63. doi: http://dx.doi.org/10.1016/j.omega. 2014.02.001

Small, M. L. (2011). How to conduct a mixed methods study: Recent trends in a rapidly growing literature. Sociology, 37(1), 57.

Sol, H. G. (1987). Conflicting experiences with DSS. Decision Support Systems, 3(3), 203-211. doi: http://dx.doi.org/10.1016/0167-9236(87)90175-8

Sörbom, D. (1989). Model modification. Psychometrika, 54(3), 371-384. doi: 10.1007/BF02294623

Spewak, S. H., & Hill, S. C. (1993). Enterprise Architecture Planning: Developing a blueprint for data, applications, and technology. Wellesley, MA, USA: QED Information Sciences, Inc.

Srivastava, S. C., & Teo, T. S. (2010). E-government, e-business, and national economic performance. Communications of the Association for Information Systems, 26(1), 14.

Stelzer, D. (2010). Enterprise Architecture Principles: Literature Review and Research Directions Service-Oriented Computing. ICSOC/ServiceWave 2009 Workshops. In A. Dan, F. Gittler, & F. Toumani (Eds.), (Vol. 6275, pp. 12-21): Springer Berlin / Heidelberg.

Stromeyer, W. R., Miller, J. W., Sriramachandramurthy, R., & DeMartino, R. (2014). The Prowess and Pitfalls of Bayesian Structural Equation Modeling: Important Considerations for Management Research. Journal of Management. doi: 10.1177/0149206314551962

Susanto, T. D. (2012). Individual Acceptance of SMS-Based e-Government Services. (Doctor of Philosophy), Flinders University, Adelaide, Australia.

Susanto, T. D., & Goodwin, R. (2013). User acceptance of SMS-based e-government services: Differences between adopters and non-adopters. Government Information Quarterly, 30(4), 486-497. doi: http://dx.doi.org/10.1016/j.giq.2013.05.010

Susilo, U. B. (n.d.). An analysis towards agencification policy in Indonesia: financial management system in Badan Layanan Umum (Public Service Agency) and the constraints in its implementation. Retrieved 8 May, 2015, from http://www.ppkblu. depkeu.go.id/index.php/baca/artikel/93/an-analysis-towards-agencification-policy-in-indonesia-financial-management-system-in-badan-layanan-umum-public-service-agency-and-the-constraints-in-its-implementation

Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics (5th ed.). Boston, MA: Pearson Education. Inc.

224

Tamm, T., Seddon, P. B., Shanks, G., & Reynolds, P. (2011). How Does Enterprise Architecture Add Value to Organisations? Communications of the Association for Information Systems, 28(1), 10.

Tat-Kei Ho, A. (2002). Reinventing Local Governments and the E-Government Initiative. Public Administration Review, 62(4), 434-444. doi: 10.1111/0033-3352.00197

The Central Intelligence Agency. (2014). The World Fact Book: Indonesia. Retrieved 1 September, 2014, from https://www.cia.gov/library/publications/the-world-factbook/geos/id.html

The Indonesian Ministry of Finance. (2011). About SPAN. Retrieved 3 October, 2011, from http://www.span.depkeu.go.id/tentang-span

The Indonesian Treasury. (2007). Vision and Mission of the Indonesian Treasury. Retrieved 30 August, 2012, from http://www.perbendaharaan.go.id/new/?pilih=hal&id=7

The International Telecommunication Union. (2011). ICT Facts and Figures. Retrieved 26 June, 2013, from http://www.itu.int/en/ITU-D/Statistics/Documents/facts/ ICTFactsFigures2011.pdf

The Open Group. (2009). TOGAF Version 9. Retrieved 9 September, 2011, from http://www.opengroup.org/togaf/

The United Nations. (2003). Global e-Government Survey 2003: E-government at the Crossroads. Retrieved 2 October, 2012, from http://unpan1.un.org/intradoc/ groups/public/documents/ un/unpan016066.pdf

The United Nations. (2004). Global e-Government Readiness Report: Toward Access for Opportunity. Retrieved 2 October, 2012, from http://unpan1.un.org/intradoc/ groups/public/documents/un/unpan019207.pdf

The United Nations. (2005). Global e-Government Readiness Report 2005: From e-Government to e-Inclusion. Retrieved 2 October, 2012, from http://unpan1.un.org/ intradoc/groups/public/ documents/un/unpan021888.pdf

The United Nations. (2008). UN e-Government Survey 2008: From e-Government to Connected Governance. Retrieved 2 October, 2012, from http://unpan1.un.org/ intradoc/groups/public/documents/un/unpan028607.pdf

The United Nations. (2010). United Nations E-Government Survey 2010: Leveraging e-government at a time of financial and economic crisis. Retrieved 2 October, 2012, from http://unpan1.un.org/intradoc/groups/public/documents/un/unpan038851. pdf

The United Nations. (2012). e-Government Survey 2012: e-Government for the People. Retrieved 2 October, 2012, from http://unpan1.un.org/intradoc/groups/public/ documents/un/unpan048065.pdf

The United Nations. (2014). e-Government Survey 2014: e-Government for the Future We Want. Retrieved 8 February, 2015, from http://unpan3.un.org/egovkb/Portals/ egovkb/Documents/un/2014-Survey/E-Gov_Complete_Survey-2014.pdf

The World Bank. (2014, 6 Nov 2014). World Development Indicators from 1960 - 2013. Retrieved 25 Nov, 2014, from http://data.worldbank.org/data-catalog/world-development-indicators

The Worldbank. (2011). Definition of E-Government. Retrieved 19 October, 2012, from http://web.worldbank.org/WBSITE/EXTERNAL/TOPICS/EXTINFORMATIONANDCOMMUNICATIONANDTECHNOLOGIES/EXTEGOVERNMENT/0,,contentMDK:20507153~menuPK:702592~pagePK:148956~piPK:216618~theSitePK:702586,00.html

Tondeur, J., Van Braak, J., & Valcke, M. (2007). Curricula and the use of ICT in education: Two worlds apart? British Journal of Educational Technology, 38(6), 962-976. doi: 10.1111/j.1467-8535.2006.00680.x

225

Torres, L., Pina, V., & Royo, S. (2005). E-government and the transformation of public administrations in EU countries: Beyond NPM or just a second wave of reforms? Online Information Review, 29(5), 531-553.

Townson, S. (2008). Why does Enterprise Architecture Matter? The Open Group White Paper. Retrieved 9 September, 2011, from http://www.opengroup.org/architecture/wp/

Trimi, S., & Sheng, H. (2008). Emerging trends in M-government. Commun. ACM, 51(5), 53-58. doi: 10.1145/1342327.1342338

Trompenaars, F., & Woolliams, P. (2011). Lost in Translation: Ignorance of differing cultural reactions to failure can stunt your company’s growth. Retrieved 10 Dec, 2013, from http://hbr.org/2011/04/lost-in-translation/ar/1

Tuunanen, T., & Kuo, I. T. (2015). The effect of culture on requirements: a value-based view of prioritization. European Journal of Information Systems, 24(3), 295-313.

U.S. Congress. (1996). Information Technology Management Reform Act (The Clinger/Cohen Act). Retrieved 7 Nov, 2011, from http://159.142.166.204/Documents/it_ management_ reform_act_feb_1996.html

Vaismoradi, M., Turunen, H., & Bondas, T. (2013). Content analysis and thematic analysis: Implications for conducting a qualitative descriptive study. Nursing & Health Sciences, 15(3), 398-405. doi: 10.1111/nhs.12048

Van Dinther, C. (2008). Agent-Based Simulation for Research in Economics. In D. Seese, C. Weinhardt, & F. Schlottmann (Eds.), Handbook on Information Technology in Finance (pp. 421-442). Heidelberg: Springer-Verlag.

Van Dyke, T. P., Kappelman, L. A., & Prybutok, V. R. (1997). Measuring information systems service quality: concerns on the use of the SERVQUAL questionnaire. MIS Quarterly, 21(2), 195-208. doi: 10.2307/249419

Van Rijckeghem, C., & Weder, B. (2001). Bureaucratic corruption and the rate of temptation: do wages in the civil service affect corruption, and by how much? Journal of Development Economics, 65(2), 307-331. doi: http://dx.doi.org/10.1016/S0304-3878(01)00139-0

van Steenbergen, M., Foorthuis, R., Mushkudiani, N., Bruls, W., Brinkkemper, S., & Bos, R. (2011, 29 Aug - 2 Sep). Achieving Enterprise Architecture Benefits: What Makes the Difference? Paper presented at the 15th IEEE International Enterprise Distributed Object Computing Conference Workshops (EDOCW), 2011.

Venkatesh, V., Bala, H., Venkatraman, S., & Bates, J. (2007). Enterprise architecture maturity: the story of the veterans health administration. MIS Quarterly Executive, 6(2), 79-90.

Venkatesh, V., Brown, S. A., & Bala, H. (2013). Bridging the qualitative-quantitative divide: Guidelines for conducting mixed methods research in information systems. MIS Quarterly, 37(1), 21-54.

Venkatesh, V., & Davis, F. D. (2000). A theoretical extension of the technology acceptance model: Four longitudinal field studies. Management Science, 46(2), 186-204.

Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User acceptance of information technology: Toward a unified view. MIS Quarterly, 27(3), 425-478.

Venkatraman, N., Henderson, J. C., & Oldach, S. (1993). Continuous strategic alignment: exploiting information technology capabilities for competitive success. European Management Journal, 11(2), 139-149.

Verdegem, P., & Verleye, G. (2009). User-centered E-Government in practice: A comprehensive model for measuring user satisfaction. Government Information Quarterly, 26(3), 487-497. doi: 10.1016/j.giq.2009.03.005

Walsham, G. (2002). Cross-cultural software production and use: a structurational analysis. MIS Quarterly, 26(4), 359-380.

226

Walsham, G., & Sahay, S. (2006). Research on information systems in developing countries: Current landscape and future prospects. Information Technology for Development, 12(1), 7-24.

Watson, H. J., & Wixom, B. H. (2007). The Current State of Business Intelligence. Computer, 40(9), 96-99. doi: 10.1109/MC.2007.331

Watson, R. T., Pitt, L. F., & Kavan, C. B. (1998). Measuring Information Systems Service Quality: Lessons from two longitudinal case studies. MIS Quarterly, 22(1), 61-79.

Wegmann, A., Balabko, P., Lê, L. S., Regev, G., & Rychkova, I. (2005). A method and tool for business-IT alignment in enterprise architecture. Paper presented at the Proceeding of the CAiSE'05 Forum, Portugal.

Welch, E. W., & Pandey, S. K. (2007). E-government and bureaucracy: Toward a better understanding of intranet implementation and its effect on red tape. Journal of Public Administration Research and Theory, 17(3), 379-404.

Wescott, C. G. (2001). E‐Government in the Asia‐pacific region. Asian Journal of Political Science, 9(2), 1-24. doi: 10.1080/02185370108434189

West, D. M. (2004). E-Government and the Transformation of Service Delivery and Citizen Attitudes. Public Administration Review, 64(1), 15-27. doi: 10.1111/j.1540-6210.2004.00343.x

Wheeldon, J., & Ahlberg, M. K. (2011). Visualizing Social Science Research: Maps, Methods, & Meaning. Thousand Oaks, CA: SAGE Publications, Inc.

Whittaker, T. A. (2012). Using the Modification Index and Standardized Expected Parameter Change for Model Modification. The Journal of Experimental Education, 80(1), 26-44. doi: 10.1080/00220973.2010.531299

Wilkinson, M. (2006). Designing an ‘adaptive’ enterprise architecture. BT Technology Journal, 24(4), 81-92. doi: 10.1007/s10550-006-0099-5

Wong, W., & Welch, E. (2004). Does E-Government Promote Accountability? A Comparative Analysis of Website Openness and Government Accountability. Governance, 17(2), 275-297. doi: 10.1111/j.1468-0491.2004.00246.x

Wu, R. C. Y. (2007). Enterprise integration in e-government. Transforming Government: People, Process and Policy, 1(1), 89-99.

Wu, Z., Christensen, D., Li, M., & Wang, Q. (2006). A Survey of CMM/CMMI Implementation in China. In M. Li, B. Boehm, & L. Osterweil (Eds.), Unifying the Software Process Spectrum (Vol. 3840, pp. 507-520): Springer Berlin Heidelberg.

Yildiz, M. (2007). E-government research: Reviewing the literature, limitations, and ways forward. Government Information Quarterly, 24(3), 646-665. doi: 10.1016/j.giq.2007.01.002

Yin, R. K. (2009). Case study research: Design and methods (4 ed. Vol. 5). Thousand Oaks, CA, USA: Sage publications.

Yin, R. K. (2011). Qualitative research from start to finish. New York, NY: Guilford Press. Yu, J., Williams, E., Ju, M., & Yang, Y. (2010). Forecasting global generation of obsolete

personal computers. Environmental science & technology, 44(9), 3232-3237. Zachman, J. A. (1987). A framework for information systems architecture. IBM Systems

Journal, 26(3), 454-470. Zachman, J. A. (1997). Enterprise architecture: The issue of the century. Database

Programming and Design, 10(3), 44-53. Zachman, J. A. (2003). The Zachman Framework for Enterprise Architecture : A Primer for

Enterprise Engineering and Manufacturing. Retrieved 5 July, 2011, from http://www. zachmaninternational.com

Zachman, J. A., & Sowa, J. F. (1992). Extending and Formalizing the Framework for Information Systems Architecture. IBM Systems Journal, 31(3), 590-616.

227

Zhao, X., Lynch, J. G., & Chen, Q. (2010). Reconsidering Baron and Kenny: Myths and truths about mediation analysis. Journal of consumer research, 37(2), 197-206.

228

ETHICS APPROVAL, LETTER OF INTRODUCTION, INVITATION TO RESEARCH, AND CONSENT FORM

FIN AL APPROV AL NOTICE

Project No.: 6205

Project Title: Improving the Indonesian Treasury e-Government Performance

Principal

Researcher: Mr Mochamad Ali Hanafiah

Email: [email protected]

Address: GPO Box 2100

Adelaide SA 5001

Approval Date: 15 August 2013 Ethics Approval Expiry Date: 31 July 2015

The above proposed project has been approved on the basis of the information

contained in the application, its attachments and the information subsequently

provided.

RESPONSIBILITIES OF RESEARCHERS AND SUPERVISORS

1. Participant Documentation Please note that it is the responsibility of researchers and supervisors, in the case of student projects, to ensure that:

mailto:[email protected]

229

all participant documents are checked for spelling, grammatical, numbering and formatting errors. The Committee does not accept any responsibility for the above mentioned errors.

the Flinders University logo is included on all participant documentation (e.g., letters of Introduction, information Sheets, consent forms, debriefing information and questionnaires – with the exception of purchased research tools) and the current Flinders University letterhead is included in the header of all letters of introduction. The Flinders University international logo/letterhead should be used and documentation should contain international dialling codes for all telephone and fax numbers listed for all research to be conducted overseas.

the SBREC contact details, listed below, are included in the footer of all letters of introduction and information sheets. This research project has been approved by the Flinders University Social and Behavioural Research Ethics Committee (Project Number ‘INSERT PROJECT No. here following approval’). For more information regarding ethical approval of the project the Executive Officer of the Committee can be contacted by telephone on 8201 3116, by fax on 8201 2035 or by email [email protected].

2. Annual Progress / Final Reports In order to comply with the monitoring requirements of the National Statement on

Ethical Conduct in Human Research (March 2007) an annual progress report

must be submitted each year on the 15 August (approval anniversary date) for

the duration of the ethics approval using the annual / final report pro forma

available from Annual / Final Reports SBREC web page. Please retain this notice

for reference when completing annual progress or final reports.

If the project is completed before ethics approval has expired please ensure a final report is submitted immediately. If ethics approval for your project expires please submit either (1) a final report; or (2) an extension of time request and an annual report. Student Projects The SBREC recommends that current ethics approval is maintained until a student’s thesis has been submitted, reviewed and approved. This is to protect the student in the event that reviewers recommend some changes that may include the collection of additional participant data. Your first report is due on 15 August 2014 or on completion of the project,

whichever is the earliest.

3. Modifications to Project Modifications to the project must not proceed until approval has been obtained from the Ethics Committee. Such matters include:

proposed changes to the research protocol;

proposed changes to participant recruitment methods;

amendments to participant documentation and/or research tools;

change of project title;

extension of ethics approval expiry date; and

changes to the research team (addition, removals, supervisor changes).


http://www.flinders.edu.au/research/info-for-researchers/ethics/committees/social-and-behavioural-research-ethics-committee/annual-and-final-reports.cfm

230

To notify the Committee of any proposed modifications to the project please submit a Modification Request Form to the Executive Officer. Download the form from the website every time a new modification request is submitted to ensure that the most recent form is used. Please note that extension of time requests should be submitted prior to the Ethics Approval Expiry Date listed on this notice.

Change of Contact Details

Please ensure that you notify the Committee if either your mailing or email address changes to ensure that correspondence relating to this project can be sent to you. A modification request is not required to change your contact details.

4. Adverse Events and/or Complaints Researchers should advise the Executive Officer of the Ethics Committee on 08

8201-3116 or [email protected] immediately if:

any complaints regarding the research are received;

a serious or unexpected adverse event occurs that effects participants;

an unforseen event occurs that may affect the ethical acceptability of the project.

http://www.flinders.edu.au/research/info-for-researchers/ethics/committees/social-and-behavioural-research-ethics-committee/modifying-an-approved-project.cfm


231

232

233

234

235

236

237

238

239

240

QUESTIONNAIRE

241

242

243

244

245

246

247

248

249

LIST OF INTERVIEW QUESTIONS

250

251

LIST OF REGULATIONS

No. Regulation Number

Date Purpose

1 260/KMK.01/2009 24/07/2009 Policy for Management of Information and Communication Technology in the Ministry of Finance

2 512/KMK.01/2009 28/12/2009 Policies and Standards for Password Usage, Electronic Mail, and the Internet in the Ministry of Finance

3 274/KMK.01/2010 24/06/2010 Policies and Standards for Electronic Data Exchange in the Ministry of Finance

4 350/KMK.01/2010 27/08/2010 Policies and Standards for Electronic Data Management in the Ministry of Finance

5 479/KMK.01/2010 13/12/2010 Policies and Standards for Information Security Management System within the Ministry of Finance

6 330/KMK.01/2011 4/10/2011 Policy and Project Management Standards for Information and Communication Technology in the Ministry of Finance

7 351/KMK.01/2011 25/10/2011 the Policy and Standard of Information Systems Development Cycle within the Ministry of Finance

8 414/KMK.01/2011 9/12/2011 Policies and Standards for Service Management Areas of Information and Communication Technology Service Support in the Ministry of Finance

9 64/KMK.01/2012 1/03/2012 Policies and Standards for Service Management Information and Communication Technology Service Delivery Areas in the Ministry of Finance

10 129/KMK.01/2012 30/04/2012 Integration of Information and Communication Technology Devices in the Ministry of Finance

11 337/KMK.01/2012 18/10/2012 Use of Domain Names in the Ministry of Finance

12 338/KMK.01/2012 18/10/2012 Direction of the Information and Communication Technology development in the Ministry of Finance

13 83/KMK.01/2013 18/03/2013 Management Policy of Information and Communication Technology Services Based on ISO 20000:2011

14 KEP-10/PB/2008 16/01/2008 Standard of Hardware and Software

252

SQL QUERY SYNTAX

The Indonesian Treasury identified its mailing administration systems database

by year. The study replaced the year in the syntax accordingly.

SELECT DISTINCT LEFT

(p.posUnit,1),u.unitNama, Count(i.mailId)

FROM

djpb_year.mail_in AS i

INNER JOIN djpb_year.mail_inpos AS p ON i.mailId = p.posId

INNER JOIN djpb_year.mail_unit AS u ON LEFT (p.posUnit, 1) = u.unitId

WHERE

i.mailSubject LIKE '%criteria%'

GROUP BY LEFT (p.posUnit, 1)

253

EXTRACTS FROM MAILING ADMINISTRATION

SYSTEMS

The Indonesian Treasury continuously evolves over time. As such, the

Indonesian Treasury reorganises its organisational structures. Some organisations have

no value in some columns as seen in Table A. 1, Table A. 2 and Table A. 3 below

mean they no longer existed in the Indonesian Treasury headquarters.

Table A. 1. The Number of Request for Reports

Unit Nama Unit 2006 2007 2008 2009 2010 2011 2012 2013

0 Direktorat Jenderal Perbendaharaan 19 28 39 51 66 51 82 85

1 Sekretariat Direktorat Jenderal Perbendaharaan 43 102 127 178 325 298 610 562

2 Direktorat Pelaksanaan Anggaran 19 12 6 17 19 7 19 14

3 Direktorat Pengelolaan Kas Negara 8 39 36 48 274 25 38 28

4 Direktorat Pengelolaan Barang Milik/ Kekayaan Negara

18

5 Direktorat Pengelolaan Surat Utang Negara

6 Direktorat Pengelolaan Pinjaman Dan Hibah Luar Negeri

9 3

7 Direkorat Pengelolaan Penerusan Pinjaman 7 64

8 Direktorat Informasi Dan Akuntansi 20 11

9 Direktorat Pengelolaan Dana Investasi

10 Direktorat Pembinaan Pengelolaan Keuangan Badan Layanan Umum

8 5 8 15 24 14

11 Direktorat Akuntansi Dan Pelaporan Keuangan 89 42 81 58 89 91

12 Direktorat Sistem Perbendaharaan 16 30 24 9 32 40

13 Direktorat Transformasi Perbendaharaan 4 3 13 12 17

14 Direktorat Sistem Manajemen Investasi 8 14 17 49 65

15 Lain-lain

Total 143 259 321 383 814 493 955 916

Table A. 2. The Number of Request for Data

Unit Unit Name 2006 2007 2008 2009 2010 2011 2012 2013






47

5 Direktorat Pengelolaan Surat Utang Negara 4


83 5





8 41 18 19 43 36





15 Lain-lain 1

Total 1122 1204 1588 2693 3066 3000 4697 4036

254

Table A. 3. The Number of Request for Data or Report Reconcilation

Unit Nama Unit 2006 2007 2008 2009 2010 2011 2012 2013






17 8

5 Direktorat Pengelolaan Surat Utang Negara


33 4





5 32 47 285 253





15 Lain-lain 1

Total 1158 1814 1629 4564 6503 4846 6378 6326

255

THE INDONESIAN TREASURY ORGANISATION

The Indonesian Treasury is an organisation under the Ministry of Finance and is

led by an Echelon 1 Director General. As of 2014, the Director General of the

Indonesian Treasury is directly assisted by the following Echelon 2 staff: one

secretary, seven directors, one treasury analyst and thirty three heads of regional

offices. Each Echelon 2 is assisted by at least five Echelon 3 staff. Each Echelon 3 is

assisted by at least three Echelon 4. Each Echelon 4 is assisted by at least two staff.

The Indonesian Treasury runs its business through its headquarters, 33 Treasury

Regional Offices (TROs) and 177 Treasury Service Offices (TSOs). Most TSOs are

located on the islands of Java, Sumatera and Borneo.

Figure A. 1. Job Rank Distribution

Echelon II, 29 Echelon III, 354 Echelon IV,

1421

ICT Engineer, 23

Staffs, 6209

Job Rank Distributions

Echelon II Echelon III Echelon IV ICT Engineer Staffs

256

Figure A. 2. The Indonesian Treasury Staff Age

Figure A. 3. Organisational Structure of Treasury Regional Office

633

1561 12022102

169

401654

1315

2 0 - 2 9 3 0 - 3 9 4 0 - 4 9 5 0 - 5 9

STAFF AGE

Male Female

Head of Treasury Regional Offices

General AffairBudget Execution

(2 Divisions) Accounting and

ReportingTSO supervision and compliance

Business Process Supervision

System Application Supervision

Internal Compliance

257

Figure A. 4. Organisational Structure of Treasury Service Office Type A1

Figure A. 5. Organisational Structure of Treasury Service Office Type A2

Head of Treasury Service Office

General Affair Spending AuthorityManagement of

Spending Units and Internal Compliance

ICT Supervisor

Bank Account Management

Accounting and Verification

Head of Treasury Service Office

General AffairSpending Authority and Management of

Spending Units

ICT Supervisor

Bank Account Management

Accounting, Verification and

Internal Compliance

258

THE INDONESIAN TREASURY E-GOVERNMENT

SYSTEMS BASIC CONFIGURATION

Table A. 4. Main Systems Used for Treasury Function

No Systems Name

System’s Description Software Database User(s)

1 AKLAP Accounting and Reporting Ms. FoxPro Oracle TRO

2 BENDUM Treasury book keeping Ms. FoxPro Ms. FoxPro TSO

3 BUKU MERAH

Government Managerial Report

Ms. FoxPro MySQL TSO

4 DIPA Budget allocation and allotment

Ms. FoxPro Ms. FoxPro TRO,TSO,SU,HQ

5 e-KEDA Fund Needs Systems Ms. FoxPro Ms. FoxPro TSO,HQ

6 e-PAYPOINT Expenses Systems Ms. FoxPro Ms. FoxPro TSO,HQ

7 GAJI Salary Ms. FoxPro MySQL SU, TSO

8 INTERFACE Decoding Revenue data Ms. FoxPro Ms. FoxPro TRO

9 LKPP Government Accounting Report

Ms. FoxPro Oracle HQ

10 PERAN Estimating budget execution Ms. FoxPro Ms. FoxPro TRO,SU

11 PERSEDIAAN Inventory Ms. FoxPro MySQL SU

12 RDI Government Investment Ms. FoxPro Ms. FoxPro HQ

13 REKSUS Special Account Transaction Systems

Ms. FoxPro Ms. FoxPro TSO,HQ

14 R-KUN Government Accounts Type 2 Ms. FoxPro Ms. FoxPro HQ

15 RPL K/L Government Accounts Type 3 Ms. FoxPro Ms. FoxPro HQ

16 RPLBUN Government Accounts Type 1 Ms. FoxPro Ms. FoxPro HQ

17 SAKPA Spending Unit (SU) Financial Accounting Systems

Ms. FoxPro Ms. FoxPro SU

18 SAPPAW SU Financial Accounting Systems at Regional Level


19 SAPPAE1 SU Financial Accounting Systems at Echelon 1 Level


20 SAPA SU Financial Accounting Systems at the Ministry Level


21 SAU Government General Accounting Systems

Ms. FoxPro Ms. FoxPro HQ

22 SAKUN Government Cash Accounting Systems

Ms. FoxPro Ms. FoxPro HQ

23 SIMAK-BMN Assets accounting Ms. FoxPro MySQL SU

24 SP2D Spending authority Ms. FoxPro MySQL TRO

25 SPM Spending order Ms. FoxPro Ms. FoxPro SU

26 SPM-PP Return Spending Order Ms. FoxPro Ms. FoxPro TSO

27 VERA Accounting and Verification Ms. FoxPro Ms. FoxPro TSO

28 WEB Transfer

Data Transfer and Monitoring PhP MySQL HQ

259

LIN

E M

INIS

TR

Y E

CH

ELO

N 1

LIN

E M

INIS

TR

Y

TRO e-GOVERNMENT SYSTEMS TSO e-GOVERNMENT SYSTEMS

BA

NK

ING

SY

ST

EM

S

THE PRESIDENT, SUPREME AUDIT BOARDS, PEOPLE REPRESENTATIVES

DIPA

LKPPAccounting Data Reconciliation

REKSUS

PERAN E-KEDAWEB

TRANSFER

MAINDB

RDI

RPL BUN

RPL K/L

BUKU MERAH

R-KUN

VERA-BUN

E-PAYPOINT

Directorate of Budgeting

Directorate of Accounting

Directorate of Cash Managment

Directorate of Treasury SystemsDirectorate of

Investment

e-Government Systems at the Indonesian Treasury Headquarters

= Accounting systems = Bank Statements = Expenses order = Data Flow

260

DIPA

AKLAP

TSO e-GOVERNMENT SYSTEMS

HEADQUARTER e-GOVERNMENT SYSTEMS

LIN

E M

INIS

TR

Y

RE

GIO

NA

L O

FF

ICE

S

Accounting Data Reconciliation

e-Government Systems at the Treasury Regional Offices

= Accounting systems

261

SP2D

REKSUS

BENDUM

GAJI

SPM-PP

E-KEDA

INTERFACEVERA

PERAN

BA

NK

ING

SY

ST

EM

S

SP

EN

DIN

G U

NIT

SY

ST

EM

S

EXPENSES

TRO e-GOVERNMENT SYSTEMS HEADQUARTER e-GOVERNMENT SYSTEMS

REVENUEDATA

BANKSTATMENTS

ACCOUNTING DATA RECONCILIATION

e-Government Systems at the Treasury Service Offices


FrontOffice

Middle Office Back Office

262

PERSEDIAAN

SIMAKBMN

SIMAKBMN

SAKPA

SPM

PERAN

DIPA

GAJI

TSO

E-GOVERNMENT SYSTEMS

SAPPAW

TRO


SIMAKBMN SAPPAE1

SIMAKBMN SAPA

HEADQUARTERS


DIPA

DIPA

SP

EN

DIN

G U

NIT

SLIN

E M

INIS

TR

Y S

RE

GIO

NA

L O

FF

ICE

S

LIN

E M

INIS

TR

Y S

EC

HE

LO

N 1

LIN

E M

INIS

TR

Y


Treasury Related e-Government Systems

at the Spending Units to Line Ministry

263

DESCRIPTIVE STATISTICS OF THE OBSERVED VARIABLES


EA1 EA2 EA3 EA4 EA5 EA6 EA7 EA8 EA9

N Valid 546 558 541 556 551 557 541 558 519

Missing 15 3 20 5 10 4 20 3 42

Mean 4.05 3.62 2.53 2.46 2.69 2.82 3.77 2.46 2.45

Median 4 4 2 2 2 2 4 2 2

Mode 5 5 2 2 2 2 5 2 2

Std. Deviation

1.278 1.295 1.062 0.922 1.09 1.235 1.35 0.891 1.137

Skewness -0.452 -0.071 1.094 0.967 1.011 1 -0.042 0.964 1.183

Kurtosis -0.644 -1.041 0.797 1.503 0.405 0.031 -1.061 1.229 1.03


OA1 OA2 OA3 OA4 OA5

N Valid 544 547 550 546 548

Missing 17 14 11 15 13

Mean 2.01 2.15 2.54 2.28 1.74

Median 2 2 2 2 2

Mode 2 2 2 2 2

Std. Deviation

0.776 0.889 0.979 0.991 0.641

Skewness 1.26 1.428 0.951 1.324 0.709

Kurtosis 3.263 3.482 0.896 2.378 1.825

Information Availability

IA1 IA2 IA3 IA4 IA5

N Valid 557 556 542 547 554

Missing 4 5 19 14 7

Mean 2.31 3.89 2.88 3.77 2.24

Median 2 4 2 4 2

Mode 2 5 2 4 2

Std. Deviation

0.84 1.36 1.309 1.287 0.904

Skewness 1.215 -0.245 0.684 -0.074 1.142

Kurtosis 2.513 -1.018 -0.563 -0.956 1.738

264

Resource Portfolio Optimisation

RP1 RP2 RP3 RP4 RP5

N Valid 544 545 557 556 552

Missing 17 16 4 5 9

Mean 2.3 2.41 1.95 2.4 3.01

Median 2 2 2 2 3

Mode 2 2 2 2 2

Std. Deviation

0.952 1.185 0.724 1.058 1.222

Skewness 1.041 1.216 1.448 1.238 0.469

Kurtosis 1.12 0.927 4.866 1.477 -0.651

Resource Complementarity

RC1 RC2 RC3 RC4 RC5

N Valid 556 547 556 558 545

Missing 5 14 5 3 16

Mean 2.16 3.66 2.48 2.09 2.49

Median 2 4 2 2 2

Mode 2 5 2 2 2

Std. Deviation

0.876 1.292 1.021 0.805 1.04

Skewness 1.134 -0.132 1.001 1.113 1.174

Kurtosis 1.758 -1.045 0.984 2.211 1.292

e-Government Performance using Initial SERVQUAL Congeneric Models

Tangible Factor

T1 T2 T3 T4

N Valid 557 561 542 551

Missing 4 0 19 10

Mean 2.39 2.24 2.52 2.4

Median 2 2 2 2

Mode 2 2 2 2

Std. Deviation

1.226 0.922 0.837 0.927

Skewness 0.966 0.775 0.819 1.05

Kurtosis 0.352 0.694 0.731 1.644

265

Reliability Factor

RL1 RL2 RL3 RL4 RL5

N Valid 561 550 561 550 560

Missing 0 11 0 11 1

Mean 2.8 2.32 1.97 2.87 2.64

Median 3 2 2 3 2.5

Mode 2 2 2 3 2

Std. Deviation

1.035 0.837 0.908 0.968 1.037

Skewness 0.533 0.62 1.142 0.494 0.726

Kurtosis -0.055 0.756 1.589 0.227 0.558

Responsiveness Factor

RS1 RS2 RS3 RS4

N Valid 536 559 551 557

Missing 25 2 10 4

Mean 2.87 2.14 2.36 3.23

Median 3 2 2 3

Mode 3 2 2 3

Std. Deviation

1.043 0.792 0.945 1.229

Skewness 0.439 1.224 0.897 0.467

Kurtosis -0.165 3.224 0.944 -0.8

Assurance Factor

A1 A2 A3 A4

N Valid 559 552 561 549

Missing 2 9 0 12

Mean 2.6 2.62 2.09 2.22

Median 2 2 2 2

Mode 2 2 2 2

Std. Deviation

0.933 0.954 0.702 0.785

Skewness 0.761 0.694 0.892 1.229

Kurtosis 0.479 0.255 2.134 2.723

266

Empathy Factor

EM1 EM2 EM3 EM4 EM5

N Valid 559 557 561 554 559

Missing 2 4 0 7 2

Mean 2.44 1.96 2.47 2.42 2.39

Median 2 2 2 2 2

Mode 2 2 2 2 2

Std. Deviation

0.807 0.685 0.892 0.788 0.929

Skewness 0.704 1.329 0.794 0.888 1

Kurtosis 0.748 4.773 0.749 1.629 0.853

267

COVARIANCE MODELS FOR VALIDITY TESTS

Figure A. 6. Initial e-Government Performance Covariance Models

268

Figure A. 7. Covariance Model for All Dimensions

269

STRUCTURAL EQUATION MODELS

E-Government Dimension

Figure A. 8. Initial e-Government Standarised Estimates

270

The Overall Model

Figure A. 9. Standarised Estimates for the Initial Model

271

Figure A. 10. Standarised Estimates for the Final Overall Model

272

The Simple Mediator Models

Figure A. 11. Standarised Estimates for the OA Dimension as Mediator

273

Figure A. 12. Standarised Estimates for the IA Dimension as Mediator

274

Figure A. 13.Standarised Estimates for the RP Dimension as Mediator

275

Figure A. 14. Standarised Estimates for the RC Dimension as Mediator

276

STATISTICAL EQUATIONS

CHI-SQUARE (χ2)

𝑋2 = ∑[(𝑂𝑖 − 𝐸𝑖)2/𝐸𝑖]

DEGREES OF FREEDOM (df)

SRMR

𝑆𝑅𝑀𝑅 =

√

(

2∑ ∑ [(

𝑆𝑖𝑗 − �̂�𝑖𝑗𝑆𝑖𝑖𝑆𝑗𝑗

)2

]𝑖𝑗=1

𝑃𝐼=1

𝑝(𝑝 + 1

)

RMSEA

𝑅𝑀𝑆𝐸𝐴 = √ 𝐹0̂𝑑

CFI

𝐶𝐹𝐼 = 1 − max(𝐹𝑚𝑖𝑛−𝑑𝑚𝑖𝑛,0)

max(𝐹𝑖−𝑑𝑖,0)

IFI

TLI